# Rankly — Full Content Index # Generated: 2026-07-20 # Source: https://www.tryrankly.com # # This file concatenates every Markdown content companion on the site. # Each page has a per-URL Markdown twin at /content/.md and is # linked from /llms.txt. This file is intended for LLM / coding-agent # consumers that prefer a single-file ingest. # # For human readers: visit the URL at the top of each section for the # rich HTML page with motion, media, and interactive elements. # Rankly | Agentic Commerce Optimization **Page URL:** https://www.tryrankly.com/ --- ## Hero **Headline:** The New Search Infra for Agents **Subtitle:** Optimize your product pages for AI Shopping Agents **CTAs:** - [Book a Demo](https://cal.com/rankly/30min) - [Learn More](/features/agentic-commerce) --- ## Optimized For (AI Platforms) Google AI Overviews, ChatGPT, Perplexity, Copilot, Gemini, Claude, Meta AI, Amazon Rufus --- ## Agent Analytics **Headline:** Measure every AI agent crawling your site, and the humans they send back See which agents visit, what they read, and what they actually send back. Measured at your edge, where every agent is visible. ### Features: - **Edge Capture** — Count every agent at the edge, even when it never runs a tag. - **Identity Checks** — Verify each bot against vendor IP ranges and auth signatures. - **Origin Mapping** — See where agents connect from across countries and networks. - **Intent Signals** — Group bots by what they do: train, index, fetch, or shop. [Read More](/features/agent-analytics) --- ## Shopping **Headline:** Track how AI recommends products and drives purchasing decisions across conversational commerce Track product visibility, retailer rankings, and pricing across AI shopping agents. ### Features: - **Shopping Responses** — See how AI recommends products across conversational shopping queries. - **Product Visibility** — Track how often your products appear in AI shopping recommendations. - **Retailer Visibility** — Track which retailers appear in AI shopping recommendations for your products. - **Pricing & Reviews** — Compare avg prices, ratings, and reviews for products across AI responses. [Read More](/features/shopping) --- ## Agentic Commerce **Headline:** AI agents are becoming the new shoppers. Make sure your products get recommended when they browse, compare, and buy ### Features: - **AI Agent Tracking** — Monitor when AI agents recommend your products across all major platforms. - **Competitive Intel** — See exactly which competitors get recommended instead of you, and why. - **Win Rate Analytics** — Track your recommendation win rate and identify optimization opportunities. - **Instant Alerts** — Get notified when competitors overtake you in AI recommendations. [Read More](/features/agentic-commerce) ### Agentic Commerce Architecture **AI Surfaces:** ChatGPT (Instant Checkout), Google AI (AI Overviews + Gemini), Perplexity (Shopping Answers), Copilot (Microsoft Shopping), Amazon Rufus (Product Discovery), AI Assistants (Third-party Agents) **Protocols:** ACP (Agentic Commerce Protocol by OpenAI), UCP (Unified Commerce Protocol by Google) **Gateway — Rankly Intelligence Agent:** Single integration point for all AI commerce channels. - Observability — Monitor agent behavior - Dynamic Catalogue — Per-channel feed optimization - Agent Control — Govern agent permissions - One API — Single integration, every platform **Your Stack:** - Answer Engine Analytics: Visibility, Citations, Query Fanouts, Sentiment, Human Traffic, Bot Traffic, Competitors, Personas, Topics, Shopping Data - Commerce Stack: Shopify, Etsy, BigCommerce, Amazon, Adobe Commerce - Your Own Analytics: PIM, OMS, Discounting, Loyalty, Catalogue, Brand Guidelines, Payments --- ## Full-Funnel AEO **Headline:** Full-funnel answer engine optimization — from discovery to conversion ### Stages: - **Discovery & Indexing** — Ensure AI engines can find, crawl, and understand your content. - **Content Optimization** — Structure content so AI models cite and recommend your brand. - **Performance Tracking** — Measure visibility, citations, and sentiment across every AI platform. - **Conversion Attribution** — Connect AI-driven traffic to revenue with end-to-end funnel analytics. ### AEO Funnel Metrics: | Stage | Value | |-------|-------| | Visibility | 78% | | Citations | 42% | | Traffic | 2,847 | | Conversion | 3.2% | | Revenue | $12.4K | [Read More](/features/answer-engine-analytics) --- ## Answer Engine Analytics **Headline:** Understand how AI search engines discover, interpret, and recommend your content to users ### Features: - **Visibility Tracking** — Monitor your brand presence across all major AI platforms. - **Prompt Testing** — Test how AI responds to queries about your products. - **Sentiment Analysis** — Measure brand perception in AI-generated responses. - **Citation Monitoring** — Track which sources AI cites for your brand. [Read More](/features/answer-engine-analytics) --- ## Traffic Analytics **Headline:** Track and analyze traffic from AI platforms with precision attribution and conversion insights ### Features: - **AI Referral Tracking** — Track traffic from ChatGPT, Perplexity, Claude & more. - **Conversion Attribution** — Compare AI-referred vs traditional search conversions. - **Journey Mapping** — Map the full AI discovery to purchase path. - **Real-time Dashboards** — Monitor AI traffic patterns as they happen. [Read More](/features/traffic-analytics) --- ## The New Ad Landscape — LLM Ads Are Here Major AI platforms are introducing advertising. Your products need to be optimized for both organic recommendations and paid placements. - **Google AI Overviews** — Live. Ads appearing within AI-generated summaries based on commercial intent. - **ChatGPT** — Testing. Sponsored product recommendations appearing inline with conversational responses. --- ## Enterprise Rankly is designed to meet the demands of modern marketing teams — secure, scalable, and ready to integrate with your existing analytics tools. **AI that amplifies your content strategy** ### Secure at Every Level Industry-grade security and compliance. Rankly uses state-of-the-art security protocols to protect your data and IP from AI misuse. ### Across Your Stack Interface and vendor agnostic. Rankly is flexible and extensible, working with any analytics provider, any CMS, and on any interface. --- ## Platform **One platform. Complete visibility.** - **Consistent Monitoring** — Track your brand mentions across all AI platforms from a single dashboard. Get alerts when your visibility changes. - **Content Optimization** — Receive AI-specific content recommendations. Understand what makes content get cited by LLMs. - **Scale Your Reach** — Expand from monitoring to optimization as you grow. Our platform scales with your needs. --- ## Pricing ### Basic — $99/month Answer Engine Analytics only. Track 5 Answer Engines, 100 Prompts, Weekly tracking, 1 account, Email Support. ### Standard — $499/month (Most Popular) AEO + Traffic + Actionables. Track 10 Answer Engines, 200 Prompts, Daily tracking, 2 accounts, Slack support. ### Enterprise — Custom Full suite with dedicated success. Track 10 Answer Engines, 400 Prompts, Daily tracking, 2 accounts, 24/7 Dedicated CS Manager. --- ## Get Started **Win every AI shopping conversation** [Start 14-day Free Trial](https://app.tryrankly.com/login) --- ## Job Openings - AI Engineer — Engineering — Remote - Data Scientist — Engineering — San Francisco, CA (New) - Frontend Engineer — Engineering — Remote - Product Manager — Product — New York, NY (New) - Growth Marketing Lead — Marketing — Remote - Enterprise Account Executive — Sales — San Francisco, CA or Remote --- ## Links - **Website:** https://www.tryrankly.com - **App:** https://app.tryrankly.com - **Book a Demo:** https://cal.com/rankly/30min - **Privacy Policy:** https://www.tryrankly.com/privacy-policy - **Terms of Service:** https://www.tryrankly.com/terms-of-service # Agentic Commerce & AI Shopping Blog | Rankly **Page URL:** https://www.tryrankly.com/blogs --- Expert insights on agentic commerce, AI shopping optimization, and generative engine strategies. Optimize your brand for ChatGPT and Gemini. --- ## Blog Posts ### How ChatGPT Search Works: The 7-Stage Pipeline Behind Every Answer **Category:** Technical | **Read Time:** 12 min | **Date:** 11 March 2026 A technical breakdown of how ChatGPT Search processes your query through 7 stages: classification, query generation, filtering, page fetch, embedding, deep-read selection, and answer generation. [Read](/blogs/how-chatgpt-search-works) --- ### News of the Week #2: The Weekly Brief on Agentic Commerce **Category:** News | **Read Time:** 5 min | **Date:** 9 March 2026 OpenAI drops Instant Checkout, Zyg raises $5.8M, Spreedly launches agentic payment rails, GenFlux raises $4.2M, and more agentic commerce news. [Read](/blogs/agentic-commerce-weekly-2) --- ### News of the Week: The Weekly Brief on Agentic Commerce **Category:** News | **Read Time:** 5 min | **Date:** 2 March 2026 Era raises $1.4M for AI agent product data, Mastercard launches Agent Pay in UAE, Walmart CEO says AI shoppers spend 35% more, and more agentic commerce news. [Read](/blogs/agentic-commerce-weekly-1) --- ### The 5 Levels of Agentic Commerce and the Missing One for Brands **Category:** Insights | **Read Time:** 7 min | **Date:** 26 February 2026 Stripe's 2025 Annual Letter outlines a framework for agentic commerce maturity. We break down the five levels and highlight the critical optimization gap most brands overlook. [Read](/blogs/agentic-commerce-levels) --- ### Inside ChatGPT's Shopping Connector: Query Fanouts Update **Category:** Product | **Read Time:** 4 min | **Date:** 12 January 2026 Our extension now captures product recommendations from ChatGPT, showing you the metadata behind each product listing. [Read](/blogs/chatgpt-shopping-connector) --- ### Query Fanouts: See What Answer Engines Look For **Category:** Product | **Read Time:** 5 min | **Date:** 26 December 2024 Discover how LLMs break down your queries into multiple search queries called query fanouts, and learn how to use them to improve your AI visibility. [Read](/blogs/query-fanouts) --- ### What is AEO? Why Generative Engine Optimization is the New SEO **Category:** Insights | **Read Time:** 6 min | **Date:** 5 November 2024 Understand how Answer Engine Optimization is reshaping digital visibility across AI answer engines. [Read](/blogs/what-is-aeo) --- ### Inside Generative Engines: A Mathematical and System-Level Breakdown **Category:** Technical | **Read Time:** 12 min | **Date:** 25 October 2024 A deep dive into the mathematical foundations and system architecture that power modern generative AI engines. [Read](/blogs/inside-generative-engines) # Agent monetization: real or still a theory? **Date:** 2026-06-23 **URL:** https://www.tryrankly.com/blogs/agent-monetization-real-or-theory **Section:** Deep Dive **Read time:** 11 min Charging AI bots to read content and agents to pay on the spot is real and live, but lopsided. The old web ran on a trade: sites let search engines crawl them, search sent readers back, and the site sold ads. AI answer engines crawl the same way but keep the answer, so the human reader never follows and the revenue dries up. Sites now want to charge the bot at the door using one old idea, the HTTP 402 Payment Required response. AWS has shipped its tollgate on Coinbase's x402 with crypto (USDC) settlement; the popular "AWS plus Stripe" line is wrong for now, Stripe support is marked coming soon. Who is paying today? Mostly the AI labs, through licensing deals, not the new tollgate: Reddit has booked $203M in content deals, OpenAI guaranteed People Inc. $16M, Perplexity set up a $42.5M publisher payout. The big transaction counts are misleading: Coinbase's x402 passed 100M but most was a crypto fad, while the one genuinely huge real deployment is Alipay in China at 300M-plus. The infrastructure is ready before the agents are, on purpose. ## The deal that broke For thirty years the web ran on a simple trade. A site let search engines crawl it for free, and in return search sent readers, who saw ads or subscribed. The crawl cost a little bandwidth and paid the site back in traffic. The arrangement held. AI answer engines broke that trade. The bot still crawls the page; the machine still shows up. What stops showing up is the human reader. The AI reads the page, writes the answer inside its own chatbot, and the person who asked gets what they needed without ever clicking through to the source. The crawl still costs the site bandwidth, but no reader follows it, so there is no ad to show and no subscriber to win. The content was served and earned nothing. This is not small. AWS says AI bots now make up more than half of web traffic for many content providers, and AI-specific crawlers grew more than 300 percent year over year. Cloudflare, which sees about a fifth of the web, recently put a date on the tipping point: bots passed humans for the first time, 57.5 percent of HTML requests against 42.5 percent (Cloudflare Radar, https://radar.cloudflare.com/traffic#bot-vs-human), a crossover its own CEO had not expected before late 2027. That has pushed publishers to an obvious question: if the bot takes the content but sends no readers back, can the bot be charged for the content? ## The fix: charge the bot at the door The fix the industry landed on is a tollgate, built on a piece of the web that has existed unused for decades. Every response carries a status code. 200 means here is the page. 404 means not found. One code was always reserved for another case: 402 Payment Required. It sat idle because there was no standard way for a machine to pay. That is the gap these systems fill. The flow is the same everywhere. A bot asks for a page. Instead of handing it over, the site replies with a 402 and a short machine-readable note: the price, and where to pay. The bot pays, the site delivers. The only real differences are what the bot pays with (crypto stablecoins in some, cards and bank payments in others) and who collects the money for the publisher. ## What AWS actually built AWS added this tollgate to its Web Application Firewall, the security layer already in front of millions of sites. A site owner turns on a Monetize rule, and when a paying-eligible bot arrives, AWS returns the 402 with a short price list. A new dashboard breaks traffic into human, search, AI, and unknown, and shows what each costs to serve. The pitch is plain: stop giving the bytes away and start charging, without touching application code. The detail most coverage gets wrong is the rail: at launch, the only way the money moves is Coinbase's x402, settling in the stablecoin USDC to a crypto wallet the content owner controls. AWS's own post says Stripe and its Machine Payments Protocol are "coming soon." So the "AWS WAF runs on Stripe" framing is a roadmap item, not the launch reality. Today it is crypto rails, which for most marketing and content teams means a crypto wallet is required to collect. What AWS got right is distribution: it made the tollgate a checkbox on infrastructure sites already run. What it has not shown is a single number, no count of publishers charging, no revenue. ## Everyone is building a version The same idea is being built by the CDNs, a payments giant, and the platforms that want agents to buy things. Split them into two groups: charging a bot to read content, and letting an agent pay for goods and services. - AWS WAF: charge bots for content at the firewall. Pays with crypto (USDC), Stripe soon. Live, Jun 2026. - Cloudflare Pay Per Crawl: block AI crawlers by default, charge to enter. Cards and bank, Cloudflare collects. Private beta, Jul 2025. - TollBit: marketplace where AI labs pay publishers. Standard billing. Live, thousands of sites. - Coinbase x402: pay-per-request for any page or API. Crypto (USDC). Live, May 2025. - Stripe MPP: let an agent pay any endpoint, built with Tempo. Crypto and cards. Live protocol, Mar 2026. - Google AP2: rules for an agent to buy on a user's behalf. Cards, bank, crypto. Live, Sep 2025, 60+ partners. - Alipay AI Pay: agents pay merchants in China. Alipay accounts. Live at scale. Two names matter most. Cloudflare sits in front of roughly a fifth of the web, so flipping its default to block AI crawlers unless they pay changed the bargaining position of every site behind it. Google's AP2, the Agent Payments Protocol, launched September 2025 with 60-plus partners including Mastercard, PayPal, Amex, Coinbase and Salesforce. AP2 is not a content toll; it is a rulebook for letting an agent buy on a user's behalf, recording what the user approved, what the agent picked, and what was charged, so a merchant can trust the purchase. It has a crypto extension built with Coinbase. Google's bet is on agents buying, which connects to the in-browser action surface in the Inside WebMCP post. There is also a fast-growing layer beneath all of this: startups whose whole job is to give an agent an identity and a wallet so it can pay at all. Skyfire issues an agent a verified identity and a spending wallet behind a "Know Your Agent" check (the agent equivalent of KYC for humans) and in late 2025 demoed an agent buying off Bose.com using Visa. Catena Labs, started by a Circle co-founder who helped create the USDC stablecoin, raised a $30M Series A in 2026 to build an AI-native bank for agents, where the human sets the spending limits and the agent moves the money. Crossmint and Nevermined sit in the same layer. These are the on-ramps that make the tollgates usable. And the toll camp is consolidating, not fragmenting. In late 2025 Cloudflare and Coinbase co-founded the x402 Foundation to make x402 the shared standard, which is why AWS, Cloudflare and Coinbase all speak the same 402 language. On the buy-side, the card networks moved too: Visa Intelligent Commerce, Mastercard Agent Pay, and PayPal checkout inside ChatGPT and Perplexity. ## So who is actually paying today? This question cuts through the hype. AI companies are paying for content today. But almost none of it flows through the new per-page tollgate. It flows through old-fashioned licensing deals, big lump sums between an AI lab and a large publisher. Real money changing hands now: - $203M: Reddit, total content licensing deals booked - $16M: OpenAI's guaranteed minimum to People Inc. - $42.5M: Perplexity's publisher revenue-share pool - $10M: Microsoft's data-access fee to Informa - $5M to $60M: per-year deals with AP, FT, News Corp, Axel Springer - ~7,000: publisher sites on the TollBit marketplace The shape of that money: a handful of large publishers signing direct deals, plus a marketplace (TollBit) gathering thousands of smaller sites and paying from hundreds to tens of thousands of dollars a month. Real revenue, but not the per-request, agent-pays-at-the-door model AWS and Cloudflare built. It is AI labs paying upfront for the right to train and answer. The summary: the tollgate is ready, but the cars are not lining up yet. The 402 toll works and is one checkbox away on AWS and Cloudflare. Today's payment volume mostly came through the older licensing door. The infrastructure was built ahead of demand, the expected pattern when every major player races to own the standard before agents arrive in force. ## The transaction numbers, and which are real The counts are real; what they measure is the question. Two are worth knowing in detail, plus the content side most coverage skips. Coinbase x402, 100M-plus transactions, mostly speculation. On-chain data from Chainalysis shows x402 going from 2,000 cumulative transactions in mid-2025 to over 100 million by mid-2026 (2K in Q2 2025, 59K in Q3, 83.9M in Q4 2025, 97.6M in Q1 2026, 100M+ in Q2 2026), with almost all of it landing after Q4 2025. The catch is in the mix: genuine micropayments (10 cents to 1 dollar, the agent-per-call range) collapsed from 46 percent of volume in early 2025 to just 4 percent by early 2026, while transactions of a dollar or more rose from 49 percent to 95 percent. The surge was led by PING, a token minted by paying 1 USDC repeatedly, which alone did more than 150,000 transactions in its first month. The typical x402 payer looks like a crypto trader, not a content-buying agent: they hold about 26 different tokens, against 4 for a normal Base wallet. A separate on-chain check put daily volume at roughly $28,000, down about 92 percent from its December 2025 peak, against a protocol ecosystem valued near $7 billion. The count is real; calling it agentic commerce is not. The content side has fewer transactions but harder leverage. Cloudflare sits in front of roughly a fifth of the web, and by its own measure AI engines return almost no traffic: earning a referral is about 750 times harder from OpenAI and roughly 30,000 times harder from Anthropic than it was from the old Google. That gap is the whole case for the toll. What Cloudflare has not published, a year after launch, is a single marketplace number, no transaction count, no revenue. Alipay AI Pay, 300M-plus transactions, the only real benchmark. In China, Ant Group's AI Pay passed 100 million users and 120 million transactions in a single week in February 2026, and 300 million-plus in total by late May, which Ant calls the first commercially scaled AI-native payment infrastructure. Real consumer use: agents ordering inside apps for Luckin Coffee, payments through Rokid smart glasses, and integration with Alibaba's Qwen models. These are Alipay's own figures, but the headline numbers were corroborated by PYMNTS. The takeaway for US and EU markets: the one place agentic payments are unambiguously at scale is China, not the West. ## Verdict: the rails are ready before the agents Is agent monetization real or theory? Real. The tollgates are live, the payment standards work, money is moving. But the build ran ahead of the demand. The per-page content toll (AWS, Cloudflare) is waiting for buyers who have not arrived in volume. The real content money today comes through old licensing deals. The huge crypto counts are mostly a fad. The one place real agent payments happen at scale is China. Even a Coinbase-backed effort admitted in March 2026 that demand "is just not there yet." That gap is the story: the infrastructure was built first on purpose, so whoever owns the standard is positioned when agents start paying at volume. ## What to do about it For content owners, the toll is a lever to hold, not pull yet. Charging a bot only works when the content is worth paying for and an agent cannot get a good-enough version free elsewhere. A few large publishers can command licensing deals; most sites cannot, and a hard paywall only makes a site invisible to the AI answers its customers now read. The safer near-term play is to be the source the agent picks, clear, structured, easy to cite, and keep the toll in reserve. The deeper point for marketing and analytics teams: this is the same decision as discovery inside answer engines, seen from the other side. A brand can be paid for access, or paid by being the answer. For almost everyone, being the answer is worth far more today. The metric that matters is not how much a bot can be charged, but whether the brand shows up when an AI answers a customer's question. ## Where the toll makes sense The pattern that works is the same in every case: leave the front door open so agents can find and cite the content, and meter the part that is genuinely scarce or expensive to produce. A few examples of how that split could play out: - News publishing: keep fresh headlines free so agents surface and attribute the story; charge for access to the deep historical archive that AI labs want for training. - Financial data: let agents read a live quote or summary at no cost; meter the bulk ingestion of historical market data that feeds models and backtests. - E-commerce: stay open to shopping agents that send buyers; price the bots harvesting full inventory and pricing for competitive intelligence. - Creative assets: index thumbnails and previews for free discovery; charge for the high-resolution originals an image or video model would train on. These are illustrations, not live products. Agent monetization is new, and we will publish more detailed articles on it in the coming days. ## Primary sources - AWS WAF adds AI traffic monetization (AWS News Blog, June 2026) - https://aws.amazon.com/blogs/aws/aws-waf-adds-ai-traffic-monetization-capability-to-help-content-owners-charge-ai-bots-for-content-access/ - Stripe and AWS WAF (Stripe Newsroom) - https://stripe.com/newsroom/news/aws-waf-and-stripe - Introducing pay per crawl (Cloudflare, July 2025) - https://blog.cloudflare.com/introducing-pay-per-crawl/ - Announcing Agent Payments Protocol AP2 (Google Cloud, September 2025) - https://cloud.google.com/blog/products/ai-machine-learning/announcing-agents-to-payments-ap2-protocol - Introducing x402 (Coinbase, May 2025) - https://www.coinbase.com/developer-platform/discover/launches/x402 - x402 agentic payments adoption, transaction counts and transfer-size distribution (Chainalysis) - https://www.chainalysis.com/blog/x402-agentic-payments-adoption/ - x402 daily volume reality check, ~$28K/day (AInvest, on-chain analysis) - https://www.ainvest.com/news/coinbase-x402-28k-daily-volume-reality-check-2603/ - Content Independence Day, crawl-to-referral difficulty (Cloudflare) - https://blog.cloudflare.com/content-independence-day-no-ai-crawl-without-compensation/ - Bot vs human HTML traffic, bots passed humans at 57.5% (Cloudflare Radar) - https://radar.cloudflare.com/traffic#bot-vs-human - Machine Payments Protocol (Stripe and Tempo, March 2026) - https://stripe.com/blog/machine-payments-protocol - How content companies get paid, licensing deal figures (Media and the Machine) - https://mediaandthemachine.substack.com/p/reddits-new-ai-licensing-deal-shows - Alipay AI Payment exceeds 120 million transactions in one week (Ant Group, Feb 2026) - https://www.businesswire.com/news/home/20260213770962/en/Alipay-AI-Payment-Exceeds-120-Million-Transactions-in-One-Week-as-Agentic-Commerce-Accelerates-in-China - Alipay next-generation AI payment infrastructure (Ant Group, May 2026) - https://www.businesswire.com/news/home/20260526337824/en/Alipay-Launches-Next-Generation-AI-Payment-Infrastructure-Debuts-AI-Wallet-and-Token-Pay-to-Power-Agentic-Economy - Coinbase-backed AI payments protocol, demand is just not there yet (CoinDesk, March 2026) - https://www.coindesk.com/markets/2026/03/11/coinbase-backed-ai-payments-protocol-wants-to-fix-micropayment-but-demand-is-just-not-there-yet **Keywords:** agent monetization, is agent monetization real, pay per crawl, AI bot monetization, charge AI bots for content, HTTP 402 payment required, x402, Coinbase x402, x402 Foundation, AWS WAF AI traffic monetization, AWS WAF Stripe, Stripe MPP, Machine Payments Protocol, Cloudflare Pay Per Crawl, TollBit, Google AP2, Agent Payments Protocol, Skyfire, Know Your Agent, Catena Labs, agent wallet, Visa Intelligent Commerce, Mastercard Agent Pay, Alipay AI Pay, AI content licensing deals, Reddit AI licensing, Perplexity publisher revenue share, USDC micropayments, agentic payments transaction volume # The 5 Levels of Agentic Commerce and the Missing One for Brands **URL:** https://www.tryrankly.com/blogs/agentic-commerce-levels **Category:** Insights | **Read Time:** 7 min | **Date:** 26 February 2026 Stripe's 2025 Annual Letter and the gap between Infrastructure and Optimization --- ## Stripe's 2025 Annual Letter outlines five levels [Stripe's 2025 Annual Letter](https://assets.stripeassets.com/fzn2n1nzq965/3LlGw839Q6kUwxZlLZDtH6/75ddcbada4aa7743dd8ec7d0f9ca497e/Stripe-annual-letter-2025-desktop.pdf) outlines five levels. --- ## The Five Levels Level 1 is eliminating web forms. You decide what to buy. The agent just fills in payment and shipping details. Level 2 is descriptive search. You stop searching for products or specific attributes and start describing situations. > "I need back-to-school supplies for a third grader in Chicago, including clothes (nothing too itchy or tight!), pencils, notebooks, and a lunch box. My son likes KPop Demon Hunters and tennis. School starts in late August." Level 3 is persistence. You stop reintroducing yourself. > "Find me options for back-to-school clothes for Bobby." Level 4 is delegation. You set constraints like budget. The agent decides and executes. Level 5 is anticipation. No prompt. The system buys before you ask. The industry is hovering between Levels 1 and 2. They compare this moment to the early web. Think Netscape, HTTP, HTML, DNS. Protocol wars. Uncertainty. An AltaVista for every Google. The standards were not obvious. The winners were not obvious. Stripe believes agentic commerce needs shared infrastructure the same way the web did. --- ## Now, the timeline of events ### September 2025 They announced the Agentic Commerce Protocol with OpenAI. ACP is an open standard for programmatic commerce between agents and businesses. ### Over the last year They introduced Shared Payment Tokens. Agents can initiate payments without exposing card credentials. Tokens are scoped by seller, time, and amount. They launched machine payments. Agents can pay for API calls and services using stablecoin micropayments. They are now collaborating with Microsoft for Copilot integrations. ### December 11, 2025 They launched the Agentic Commerce Suite. This is the practical layer. Hosted ACP endpoints. Catalog syndication. Checkout sessions. Fraud handling. One integration instead of six months per agent. Participants already onboarding include Etsy, Urban Outfitters, Anthropologie, Coach, and Kate Spade. --- ## So, what does Stripe's Agentic Toolkit actually help you with ### Discovery You get a hosted ACP endpoint. Your catalog becomes machine readable and discoverable by agents. ### Checkout Shipping, taxes, sessions, order events. You can use Stripe's stack or plug in your own. ### Payments Shared Payment Tokens reduce credential risk. Agents can transact safely. ### Fraud Radar adapts signals so agents are not misclassified as bots. New fraud patterns are monitored at the token level. --- ## What it does not solve It does not decide which of your products should be surfaced to which agent. Agents optimize for user intent signals like low price or best for X. They do not understand your unit economics. They do not know: - Some SKUs have high return rates - Some have low repeat purchase - Some have high churn - Some get damaged/stale/delayed in delivery - Some cause support tickets - Some have low margins Agents match query strings. They do not see downstream business impact unless you explicitly encode it. If you expose your full catalog blindly, you are letting external agents optimize for user's query, not your profitability. This is the gap most brands are missing. Stripe is building the rails. Protocols, tokens, checkout, fraud. But optimization across agents is your responsibility. **And this is where the real game starts.** --- ## Brand Readiness Levels The five levels of agentic commerce are about user autonomy. But brands need their own five levels of readiness: - **L1** - Basic exposure - **L2** - Clean structured catalog - **L3** - Real time inventory and pricing - **L4** - Profit aware filtering - **L5** - Cross agent optimization Most companies are still at level one. --- ## Conclusion The hype is about Level 5 anticipation. The reality is we are wiring Level 1 infrastructure. That is not a bad thing. It is how platform shifts happen. If you are interested to know more about Agentic Commerce, follow us. We keep sharing regular updates :) Also, If you are interested to partner with us, Rankly is an agentic commerce optimization platform that helps you decide what actions to take for which products and for what agents/channels. All in one single platform, without the hassle to keep up with all the agentic commerce platforms. # News of the Week: The Weekly Brief on Agentic Commerce **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-1 **Category:** News | **Read Time:** 5 min | **Date:** 2 March 2026 The weekly brief on Agentic Commerce --- ## Era Raises $1.4M Pre-Seed to Optimize Product Data for AI Agents [Era](https://techfundingnews.com/era-shopping-1-4m-chatgpt-ai-agents-exclusive/) raised $1.4M with a clear thesis: product metadata optimized for AI agents is the next growth lever. If your catalog isn't structured for ChatGPT, Perplexity, Gemini-style agents, you won't surface in agent-led shopping flows. --- ## Unicity Labs Launches Agentic Commerce Accelerator in India Swiss startup [Unicity Labs](https://www.prnewswire.com/in/news-releases/padup-ventures-and-unicity-labs-partner-to-bring-agentic-commerce-infrastructure-to-indiwi-302699594.html) partnered with Padup Ventures to bring agentic commerce infrastructure to Indian startups. Agent-ready commerce infra is expanding into emerging markets. --- ## Netcore Report: Agentic Commerce Becoming E-Commerce Operating System A [new report from Netcore Cloud](https://netcorecloud.com/the-agentic-commerce-shift-report-2026-trends/) frames agentic commerce as the new operating layer of online retail. Morgan Stanley projects by 2030: - 50% of online shoppers will use AI agents - 25% of online spend will flow through them Adoption is expected to become mainstream within five years. --- ## Mastercard Launches Agent Pay in UAE [Mastercard](https://europeanbusinessmagazine.com/business/agentic-commerce-when-ai-buys-on-your-behalf-who-pays-whos-liable/) rolled out Agent Pay in the UAE, with Europe next. AI agents can now complete purchases autonomously within supported ecosystems. Payments are becoming agent-executable. --- ## Walmart CEO: AI Shoppers Spend 35% More [Walmart CEO John Furner](https://avenuez.com/blog/walmarts-ai-assistant-leads-to-35-larger-order-values/) shared that customers using Walmart's AI assistant spend 35% more per order. Separately, Alipay reported 120M+ AI-driven transactions in a single week in China. AI-assisted shopping is already impacting revenue at scale. --- ## Salesforce Signs Agreement to Acquire Cimulate [Salesforce](https://brandsit.pl/en/salesforce-acquires-cimulate-agentforce-commerce-with-new-ai-engine/) signed a definitive agreement to acquire Cimulate, an AI-driven product discovery company. Enterprise commerce platforms are integrating agent-led discovery into their core stack. --- ## Alipay AI Pay Exceeds 120M Transactions in a Week [Alipay's AI Pay](https://digitalcfoasia.com/alipay-ai-transactions-surpass-120-million-in-a-single-week-as-agentic-commerce-expands-in-china/) solution, which enables payments through AI agents, exceeded 120 million transactions in a single week, marking a major usage milestone as agentic commerce accelerates in China. The system is already integrated into retail mini-programs and AI devices. # News of the Week #10: The Weekly Brief on Agentic Commerce > FIDO Alliance launches agentic working groups with Google AP2 and Mastercard Verifiable Intent. Ant International open-sources AMP across 1.8B Alipay+ wallets. OKX ships its own Agent Payments Protocol. Stripe Sessions 2026 ships 288 products. Visa Agentic Ready APAC, BigCommerce + PayPal, Amazon Rufus +115% MAU, Wizard + Mastercard + Stripe, Snap AI Sponsored Snaps with Experian. **Published:** 2026-05-04 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-10 ## In this article 1. FIDO Alliance + AP2 + Verifiable Intent 2. Ant International AMP across Alipay+ 3. OKX Agent Payments Protocol 4. Stripe Sessions 2026 (288 products) 5. Visa Agentic Ready APAC (50+ partners) 6. BigCommerce + PayPal Store Sync 7. Amazon Q1: Rufus MAU +115% 8. Wizard + Mastercard + Stripe 9. Snap AI Sponsored Snaps + Experian --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-10](https://www.tryrankly.com/blogs/agentic-commerce-weekly-10). **Keywords:** Agentic Commerce, AI Shopping, FIDO Alliance, AP2, Verifiable Intent, Ant International AMP, OKX Agent Payments, Stripe Sessions, Visa Agentic Ready APAC, BigCommerce PayPal, Amazon Rufus, Wizard Mastercard, Snap AI # News of the Week #11: The Weekly Brief on Agentic Commerce > OpenAI ships self-serve ChatGPT Ads Manager and paid Product Feed Campaigns, Etsy launches a native ChatGPT app, Google drops FAQ rich results from Search (FAQPage schema still feeds AI surfaces), Visa Agentic Ready expands to Canada, Mastercard runs Portugal's first live Agent Pay, ICSC and McKinsey forecast US agentic commerce at $1T by 2030, and Sopra Steria pegs Europe at €310B. **Published:** 2026-05-12 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-11 ## In this article 1. OpenAI ChatGPT Ads Manager + Product Feed Campaigns 2. Etsy launches native ChatGPT app (100M+ listings) 3. Google drops FAQ rich results (FAQPage schema still feeds AI) 4. Visa Agentic Ready expands to Canada (all 5 major banks) 5. Mastercard opens Lisbon hub + Portugal's first live Agent Pay 6. ICSC + McKinsey: US agentic commerce projected at $1T by 2030 7. Sopra Steria: €310B European TAM, 41% don't trust AI shopping agents --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-11](https://www.tryrankly.com/blogs/agentic-commerce-weekly-11). **Keywords:** Agentic Commerce, AI Shopping, OpenAI ChatGPT Ads, ChatGPT Product Feed, Etsy ChatGPT app, Google FAQ rich results, Visa Agentic Ready Canada, Mastercard Lisbon, ICSC McKinsey $1T, Sopra Steria # News of the Week #12: The Weekly Brief on Agentic Commerce > Amazon retires Rufus and ships Alexa for Shopping to the main search bar with conditional-purchase rules and an expanded Buy for Me agent. OpenAI launches Personal Finance in ChatGPT and wires Plaid into 12,000+ banks. Google + Affirm + Klarna make BNPL native to AI Mode and Gemini via the Universal Commerce Protocol. Stripe's John Collison goes on Bloomberg Odd Lots. Anthropic launches Claude for Small Business and re-opens metered third-party agent tool credits with an expanded PwC partnership. Stellagent ships an Agentic Commerce Studio validating ACP, UCP, AP2, Visa TAP and Mastercard Agent Pay in one place. Coinbase x402 goes live on Arbitrum. **Published:** 2026-05-19 **Category:** News **Coverage window:** 11 to 17 May 2026 **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-12 ## In this article 1. Amazon retires Rufus and promotes an Alexa shopping agent to the main search bar 2. OpenAI launches Personal Finance in ChatGPT and wires Plaid into 12,000+ banks 3. Google + Affirm + Klarna make BNPL native to AI Mode and Gemini via UCP 4. Stripe's John Collison on Bloomberg Odd Lots: keyword search is ridiculous in 2026 5. Anthropic launches Claude for Small Business with pre-built commerce workflows 6. Anthropic re-opens metered third-party agent tools, scales PwC, ships 12 legal plugins 7. Stellagent launches a multi-protocol Agentic Commerce Studio (ACP, UCP, AP2, Visa TAP, Mastercard Agent Pay) 8. Coinbase x402 goes live on Arbitrum, taking the HTTP-402 agent rail multi-chain --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-12](https://www.tryrankly.com/blogs/agentic-commerce-weekly-12). **Keywords:** Agentic Commerce, AI Shopping, Amazon Alexa for Shopping, Amazon Rufus retired, Amazon Buy for Me agent, OpenAI Personal Finance ChatGPT, ChatGPT Plaid integration, Universal Commerce Protocol UCP, Affirm Google AI Mode BNPL, Klarna Gemini BNPL, Stripe John Collison Odd Lots, Claude for Small Business, Anthropic metered agent tool credits, Anthropic PwC expansion, Stellagent Agentic Commerce Studio, ACP UCP AP2 protocol testbed, x402 Arbitrum Coinbase # News of the Week #13: The Weekly Brief on Agentic Commerce > Google I/O ships Universal Cart, AP2 v0.2.0 (donated to FIDO), and Gemini Spark. Marketing Live adds Amazon, Meta, Microsoft, Salesforce, and Stripe to the UCP Tech Council, and Stripe and Google formalize a UCP partnership so Gemini and AI Mode checkouts clear through Stripe. Klarna launches a Shopping Search app inside ChatGPT via its Product Search MCP server. Expedia Group B2B unveils an AI toolkit and a forthcoming B2B MCP server at Explore 26 and introduces a B2A (business-to-agent) marketing thesis. Airbnb's 2026 Summer Release ships an expanded multilingual AI assistant plus horizontal expansion into hotels, car rentals, and grocery. Pattern Group (NASDAQ: PTRN) launches Pattern Intelligence with a GEO + Alexa for Shopping Scorecard across Sparky, Alexa, ChatGPT, and Google AI Mode. Fireblocks ships an Agentic Payments Suite and joins the x402 Foundation. Walmart Q1 FY27 earnings disclose 35% higher AOV from Sparky shoppers, 2x weekly active users QoQ, and 4x units. **Published:** 2026-05-26 **Category:** News **Coverage window:** 18 to 24 May 2026 **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-13 ## In this article 1. Google I/O 2026: Universal Cart, AP2 donated to FIDO, Gemini Spark 2. Google Marketing Live: Amazon, Meta, Microsoft, Salesforce, and Stripe join the UCP Tech Council; Stripe-Google UCP partnership 3. Klarna launches AI-powered Shopping Search app in ChatGPT via its Product Search MCP server 4. Expedia Group B2B unveils an AI toolkit + B2B MCP server at Explore 26, plus a formal B2A marketing thesis 5. Airbnb 2026 Summer Release: multilingual AI assistant handles 40% of queries, horizontal expansion into hotels, cars, and grocery 6. Pattern Group launches Pattern Intelligence (Pi) with a GEO + Alexa for Shopping Scorecard across Sparky, Alexa, ChatGPT, and Google AI Mode 7. Fireblocks ships an Agentic Payments Suite and joins the x402 Foundation 8. Walmart Q1 FY27 earnings: Sparky shoppers carry 35% higher AOV, weekly actives 2x, units 4x ## Primary sources - Google: Universal Cart launch (blog.google) - https://blog.google/products-and-platforms/products/shopping/google-shopping-cart/ - Google: Marketing Live shopping updates - https://blog.google/products-and-platforms/products/shopping/shopping-updates-google-marketing-live/ - Klarna Shopping Search in ChatGPT (BusinessWire) - https://www.businesswire.com/news/home/20260520024278/en/Klarna-launches-AI-powered-Shopping-Search-app-in-ChatGPT - Expedia Group B2B AI toolkit + MCP server (BusinessWire) - https://www.businesswire.com/news/home/20260519530379/en/Expedia-Group-B2B-Introduces-AI-Toolkit-and-Platform-for-the-Future-of-Travel-Distribution - Airbnb 2026 Summer Release - https://news.airbnb.com/airbnb-2026-summer-release/ - Pattern Intelligence (Pi) launch (BusinessWire) - https://www.businesswire.com/news/home/20260521129799/en/Pattern-Launches-Pattern-Intelligence-Pi-to-Accelerate-AI-Powered-Ecommerce-Growth-for-Global-Brands - Fireblocks Agentic Payments Suite + x402 Foundation (PRNewswire) - https://www.prnewswire.com/news-releases/fireblocks-joins-x402-foundation-launches-agentic-payments-suite-302777251.html - Walmart Sparky Q1 FY27 numbers (Digital Commerce 360) - https://www.digitalcommerce360.com/2026/05/22/walmart-sparky-agent-ai-sales-supply-chain/ ## Secondary coverage - Universal Cart (TechCrunch) - https://techcrunch.com/2026/05/19/googles-new-universal-cart-wants-to-follow-you-across-the-entire-internet/ - UCP Tech Council + new verticals (Search Engine Land) - https://searchengineland.com/google-expands-universal-commerce-protocol-and-launches-new-agentic-shopping-tools-478113 - Stripe-Google UCP partnership (Payments Dive) - https://www.paymentsdive.com/news/stripe-google-partner-on-agentic-commerce/818915/ - Klarna ChatGPT app (PYMNTS) - https://www.pymnts.com/artificial-intelligence-2/2026/klarna-launches-chatgpt-shopping-app-with-live-prices/ - Expedia agentic AI for B2B partners (Skift) - https://skift.com/2026/05/21/expedia-to-launch-agentic-ai-tools-for-b2b-partners/ - Airbnb Summer Release (TechCrunch) - https://techcrunch.com/2026/05/20/airbnb-gets-into-hotels-expands-ai-for-host-onboarding-and-customer-support/ - Pattern Intelligence Pi (PYMNTS) - https://www.pymnts.com/news/ecommerce/2026/pattern-group-launches-ai-powered-execution-engine-for-ecommerce/ - Fireblocks x402 commitment (TradingView / Cointelegraph) - https://www.tradingview.com/news/cointelegraph:efaa4241b094b:0-fireblocks-launches-agentic-payment-support-joins-x402-foundation/ - Walmart Sparky analysis (Inc.) - https://www.inc.com/bill-murphy-jr/walmart-just-revealed-some-fascinating-numbers-about-its-ai-shopping-agent-and-this-key-trend-mattered-most/91349677 --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-13](https://www.tryrankly.com/blogs/agentic-commerce-weekly-13). **Keywords:** Agentic Commerce, AI Shopping, Google I/O 2026 Universal Cart, AP2 FIDO Alliance, Gemini Spark, UCP Tech Council, Stripe Google UCP partnership, Klarna ChatGPT MCP, Expedia Group B2B MCP, B2A business to agent, Airbnb Summer Release 2026, Pattern Intelligence Pi, Pattern Group PTRN, GEO Alexa Shopping Scorecard, Fireblocks Agentic Payments, x402 Foundation, Walmart Sparky 35 percent AOV # News of the Week #14: The Weekly Brief on Agentic Commerce > AWS productizes Alexa for Shopping and sells it to outside retailers (Kate Spade lead customer, $12B in incremental amazon.com sales). Robinhood ships Agentic Trading and the first agent-payable Gold credit card from a major retail brand (stock +28%). Visa invests in Replit and turns the IDE into a Trusted Agent Protocol on-ramp. Highnote ships the first commercial deployment on Visa Intelligent Commerce Connect. Shopify formalizes agentic-commerce distribution in its merchant ToS. Stord raises $250M Series F at $3B + opens Stord Labs. Coinbase Base ships Base MCP for Claude / ChatGPT / Cursor. Claude Opus 4.8 lifts the web-agent benchmark that maps to shopping. Keyrock: 176M on-chain agent transactions, $73M settled, 98.6% USDC. Forrester calls a 2026 hype reset. **Published:** 2026-06-01 **Category:** News **Coverage window:** 25 to 31 May 2026 **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-14 ## In this article 1. AWS productizes Alexa for Shopping and sells it to outside retailers 2. Robinhood ships Agentic Trading + first agent-payable credit card 3. Visa invests in Replit and turns the IDE into a Trusted Agent Protocol on-ramp 4. Highnote ships the first commercial deployment on Visa Intelligent Commerce Connect 5. Shopify formalizes agentic-commerce distribution in merchant ToS 6. Stord raises $250M Series F at $3B + opens Stord Labs 7. Coinbase Base ships Base MCP for Claude, ChatGPT, and Cursor 8. Claude Opus 4.8 lifts the web-agent benchmark that maps to shopping 9. Keyrock: 176M on-chain agent transactions and $73M settled in trailing year 10. Forrester: agentic commerce hype reset for mid-2026 --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-14](https://www.tryrankly.com/blogs/agentic-commerce-weekly-14). **Keywords:** Agentic Commerce, AWS Agentic Shopping Assistant, Alexa for Shopping retailers, Kate Spade AI Gift Concierge, Robinhood Agentic Trading, Robinhood Gold agent card, Visa Replit, Trusted Agent Protocol, Highnote VICC, Shopify agentic ToS, Stord Series F, Coinbase Base MCP, Claude Opus 4.8, Keyrock USDC, Forrester hype reset # News of the Week #15: The Weekly Brief on Agentic Commerce > The week agentic rails went from announced to in production: browser-native tool exposure via WebMCP, the first live European card-network agent payment from Worldline, ING, and Mastercard, the UK's first agent checkout from Hey Savi and PayPal, and stablecoin agent settlement in travel from Travala all shipped inside 72 hours. In parallel, the AI answer became a market as ChatGPT ads went live in the UK while Google and Burson handed the AI-visibility category its measurement layer. Cloudflare confirmed bots crossed the majority of web traffic, and Amazon v. Perplexity reached the Ninth Circuit to decide whether agents have a right to visit at all. **Published:** 2026-06-09 **Category:** News **Coverage window:** 1 to 7 June 2026 **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-15 ## In this article 1. WebMCP graduates to a public origin trial in Chrome 149 2. Worldline, ING, and Mastercard complete the first live agentic payment in Europe 3. Hey Savi and PayPal launch the UK's first agentic commerce platform with in-app checkout 4. ChatGPT ads go live in the UK, the first market outside North America and ANZ 5. Google ships Generative AI performance reports in Search Console 6. Cloudflare: bot traffic crosses the majority threshold at 57.5% 7. Travala launches Travel MCP with gasless USDC settlement on Base via x402 8. Microsoft unveils seven in-house MAI models at Build 2026 9. Burson and Profound publish The Credibility Paradox 10. Nectar Social raises $30M Series A and rebrands to Nectar Agent 11. Amazon v. Perplexity reaches Ninth Circuit oral arguments --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-15](https://www.tryrankly.com/blogs/agentic-commerce-weekly-15). **Keywords:** agentic commerce, WebMCP, Mastercard Agent Pay, PayPal Agentic Commerce Services, ChatGPT ads, Google Search Console AI, Cloudflare bot traffic, Travala x402, Microsoft MAI models, Burson Credibility Paradox, Nectar Agent, Amazon v. Perplexity, generative engine optimization, USDC stablecoin payments # News of the Week #16: The Weekly Brief on Agentic Commerce > The week the payment networks built the agentic rails: Visa put its credentials inside ChatGPT through an OpenAI collaboration, Mastercard launched Agent Pay for Machines hours later, and Pine Labs shipped P3P, India's first agentic payment protocol on UPI. Checkout.com's data showed demand forming fast while trust lags, ShopAgentic raised EUR 1.9M for an agent-native store, fresh GEO research found ChatGPT and Perplexity cite almost entirely different webs, and Amazon v. Perplexity reached Ninth Circuit oral arguments over whether a user's shopping agent counts as the user under the CFAA. **Published:** 2026-06-16 **Category:** News **Coverage window:** 8 to 15 June 2026 **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-16 ## In this article 1. Visa partners with OpenAI to put its rails inside ChatGPT 2. Mastercard launches Agent Pay for Machines, hours later 3. Visa's Agent Score, Agentic Directory, and a $7B stablecoin run-rate 4. Pine Labs launches P3P, India's first agentic payment protocol on UPI 5. Checkout.com: demand is forming fast, trust is still catching up 6. ShopAgentic raises EUR 1.9M to build an agent-native store 7. GEO data: ChatGPT and Perplexity cite almost different webs 8. Amazon v. Perplexity reaches Ninth Circuit oral arguments --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-16](https://www.tryrankly.com/blogs/agentic-commerce-weekly-16). **Keywords:** agentic commerce, Visa Intelligent Commerce, OpenAI, ChatGPT, Mastercard Agent Pay for Machines, AP4M, HTTP 402, stablecoins, Pine Labs P3P, UPI, Grantex, Checkout.com, ShopAgentic, GEO, generative engine optimization, Perplexity, Amazon v. Perplexity, CFAA, Agent Score, WebMCP, Rankly # Agentic Commerce Weekly #2: AI Shopping News Roundup | Rankly **Page URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-2 **Published:** March 9, 2026 **Category:** News **Read Time:** 5 min --- OpenAI drops Instant Checkout, Zyg raises $5.8M, Spreedly launches agentic payment rails, GenFlux raises $4.2M, and more agentic commerce news. # Agentic Commerce Weekly #3: AI Shopping News Roundup | Rankly **Page URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-3 **Published:** March 16, 2026 **Category:** News **Read Time:** 5 min --- Google's UCP gains payment support, Wizard partners with Stripe on ACP, Lemrock raises €6M, Amazon blocks Perplexity's AI shopping agent, and more agentic commerce news. # News of the Week #4: The Weekly Brief on Agentic Commerce > Walmart pulls back on in-chat checkout, Stripe launches Machine Payments Protocol, Shoplazza adopts agentic architecture, Rezolve and Microsoft push agentic commerce, Adyen releases whitepaper. **Published:** 2026-03-23 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-4 ## In this article 1. Walmart In-Chat Checkout 2. Stripe Machine Payments 3. Shoplazza Agentic Architecture 4. Rezolve + Microsoft 5. Adyen Whitepaper --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-4](https://www.tryrankly.com/blogs/agentic-commerce-weekly-4). **Keywords:** Walmart, Sparky AI, Stripe MPP, Machine Payments Protocol, Shoplazza, Rezolve AI, Microsoft, Adyen, agentic commerce # News of the Week #5: The Weekly Brief on Agentic Commerce > Razorpay x Sarvam AI launch voice commerce, Firmly Connect goes live, Shopify launches Agentic Plan, OpenAI expands ACP, Stripe x Meta one-click checkout, ChatGPT ads cross $100M, and more. **Published:** 2026-03-29 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-5 ## In this article 1. Razorpay x Sarvam AI 2. Firmly Connect 3. Reddit Shopping Features 4. Shopify Agentic Plan 5. OpenAI Updates ACP 6. Accenture x DaVinci 7. Stripe x Meta Checkout 8. ChatGPT Ads $100M --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-5](https://www.tryrankly.com/blogs/agentic-commerce-weekly-5). **Keywords:** Razorpay, Sarvam AI, Firmly Connect, Shopify Agentic Plan, Reddit shopping, OpenAI ACP, Accenture DaVinci, Stripe Meta, ChatGPT ads, agentic commerce # News of the Week #6: The Weekly Brief on Agentic Commerce > Visa drops B2AI report, Visa x Ramp expand Trusted Agent Protocol, OpenAI x Smartly.io launch conversational ads, Macy's AI chatbot drives 4.75x spending, Rezolve AI hits 543% growth, and more. **Published:** 2026-04-06 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-6 ## In this article 1. Visa B2AI Report 2. Visa x Ramp 3. OpenAI x Smartly.io 4. Macy 5. Rezolve AI 543% Growth 6. Parallel Eve for Shopify 7. Juniper $1.5T Forecast 8. Stripe Shoptalk Recap 9. PwC Shoptalk Recap --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-6](https://www.tryrankly.com/blogs/agentic-commerce-weekly-6). **Keywords:** Visa B2AI, Visa Ramp, OpenAI Smartly, Macy's AI, Rezolve AI, Parallel Eve, Juniper Research, Stripe Shoptalk, PwC Shoptalk, agentic commerce # News of the Week #7: The Weekly Brief on Agentic Commerce > Visa launches Intelligent Commerce Connect, Mastercard takes Agent Pay live across Asia, Syndigo launches Synapse PXM, Forrester's agentic payments assessment, and Google's 2026 retail ad playbook. **Published:** 2026-04-13 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-7 ## In this article 1. Visa Intelligent Commerce 2. Mastercard Agent Pay Asia 3. Syndigo Launches Synapse 4. Forrester Assessment 5. Google Retail Ad Playbook --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-7](https://www.tryrankly.com/blogs/agentic-commerce-weekly-7). **Keywords:** Agentic Commerce, AI Shopping, Visa Intelligent Commerce Connect, Mastercard Agent Pay, Syndigo Synapse, Forrester, Google UCP # News of the Week #8: The Weekly Brief on Agentic Commerce > Amex launches ACE Developer Kit with agent purchase protection, Etsy reports 15x YoY agentic traffic growth, Lobster.cash integrates Mastercard Agent Pay, AEON360 goes all-in on UCP, Deloitte maps APAC's agentic future, and OpenAI readies PPC for ChatGPT Ads. **Published:** 2026-04-20 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-8 ## In this article 1. Amex ACE Developer Kit 2. Etsy: 15x Agentic Traffic 3. Lobster.cash + Mastercard 4. AEON360 + Google Cloud 5. Deloitte: APAC Drives Agentic 6. Agentic Commerce Summit 7. OpenAI Ads Manager --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-8](https://www.tryrankly.com/blogs/agentic-commerce-weekly-8). **Keywords:** Agentic Commerce, AI Shopping, American Express ACE, Etsy Agentic Traffic, Mastercard Agent Pay, AEON360 UCP, Deloitte APAC, OpenAI Ads Manager # News of the Week #9: The Weekly Brief on Agentic Commerce > Adobe ships a Commerce MCP Server with CX Enterprise, Microsoft drops four agentic commerce updates in one post, Pine Labs launches India's Agentic Commerce Suite, Ulta and Google launch Gemini-powered agentic shopping, Accenture and Google Cloud expand on Gemini Enterprise, Amex scales BigQuery and posts a VP role, Coinbase ships Agent.market on x402, and HUMAN Security adds Agentic Visibility inside Adobe Experience Platform. **Published:** 2026-04-27 **Category:** News **URL:** https://www.tryrankly.com/blogs/agentic-commerce-weekly-9 ## In this article 1. Adobe CX Enterprise + Commerce MCP Server 2. Microsoft x4 Agentic Commerce Updates 3. Pine Labs Agentic Commerce Suite (India) 4. Ulta Beauty + Google Gemini 5. Accenture + Google Cloud Gemini Enterprise Acceleration 6. American Express BigQuery scale-up + VP role 7. Salesforce, BCG, and Google Cloud on Gemini Enterprise 8. Coinbase Agent.market on x402 9. HUMAN Security Agentic Visibility in Adobe Experience Platform 10. Insignia VC: When Agents Go Shopping --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/agentic-commerce-weekly-9](https://www.tryrankly.com/blogs/agentic-commerce-weekly-9). **Keywords:** Agentic Commerce, AI Shopping, Adobe MCP, Microsoft Copilot UCP, Pine Labs, Ulta Gemini, Accenture, Amex BigQuery, Coinbase x402, HUMAN Security # Inside ChatGPT Ads: A Full XHR Breakdown of the Ad Snippet Pipeline > We intercepted ChatGPT's network logs to reverse-engineer exactly how ads are served, including the full ad unit schema, tracking endpoints, and CDN infrastructure. **Published:** 2026-04-01 **Category:** Technical **URL:** https://www.tryrankly.com/blogs/chatgpt-ad-infrastructure-breakdown ## In this article 1. What We Did 2. The Query 3. Conversation Stream 4. Product Update Endpoint 5. The Ad Snippet 6. Ad Tracking 7. Full Ad Schema 8. What This Means --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/chatgpt-ad-infrastructure-breakdown](https://www.tryrankly.com/blogs/chatgpt-ad-infrastructure-breakdown). **Keywords:** ChatGPT ads, ChatGPT ads, OpenAI advertising, XHR analysis, ad snippet schema # Inside ChatGPT's Shopping Connector: Query Fanouts Update **URL:** https://www.tryrankly.com/blogs/chatgpt-shopping-connector **Category:** Product Update | **Read Time:** 8 min | **Date:** 12 January 2026 Our extension now captures product recommendations from ChatGPT, showing you the metadata behind each product listing. *Part 2 of the Series (← [Query Fanouts Blog](/blogs/query-fanouts))* --- ## A Quick Recap In a [previous blog](/blogs/query-fanouts), we introduced Query Fanouts, the sub-queries ChatGPT generates when researching your question. Our extension captures those, giving you visibility into what the AI is actually searching for. For shopping-related questions like *"best mechanical keyboard under 10000"* or *"wireless earbuds for running"*, ChatGPT goes a step further. It doesn't just search the web; it fetches product data and displays recommendations directly in its response, complete with prices, ratings, and merchant links. We wanted to know where that product data comes from and how it gets there. So we built a feature to capture it. --- ## What's New: Product Recommendations Capture Now our extension can also capture the products listed in ChatGPT's response, alongside the query fanouts we were already capturing. In addition, we extract and display the metadata behind every product being listed in GPT's response. The products are displayed in the same carousel format as ChatGPT shows them. When you click on any product, a metadata panel appears below with additional data collected from the API response. ### Product Metadata #### Overview - **AI Rerank** - Shows the product's position in the final carousel and how much it moved from its original position - **Source Pos** - The original position and citation reference from the API response - **Rating** - Star rating and review count when available - **Listing Type** - Multi-Seller, Single in Cluster, or Standalone #### Search Context - **Market** - Country and language codes (e.g., IN / EN) - **Location** - User location if detected - **Query** - The search query sent to fetch products #### Data Sources - **Provider** - Where the product data appears to come from - **gpcid** - A product cluster identifier - **Merchant ID** - A merchant identifier when present - **Catalog ID** - A catalog reference number --- ## What We Observed We ran a handful of product queries and looked at the captured data. A few patterns stood out. ### The p2 Provider Every product we captured had a `providers: ["p2"]` tag. Based on the structure of the decoded tokens, this appears to be **Google Shopping** data. ### Carousel Order vs Internal Index Each product has an internal reference like `turn0product0`. You'd assume `product0` appears first in the carousel. It doesn't always. ChatGPT seems to reorder products after receiving them, possibly based on relevance to the specific question. ### The MID Pattern We noticed that some products have all three identifiers (catalogid, gpcid, mid), while others only have a product ID. This led us to create three labels: | Label | What We Observed | |-------|-----------------| | Multi-Seller | Has `mid`, multiple merchants shown | | Single in Cluster | Has `gpcid` but no `mid` | | Standalone | Only `productid`, single exclusive listing | --- ## Location Context Every product token we decoded contained a `uule` parameter, an encoded location string. This suggests the Shopping Connector is geo-targeted. The products you see may differ based on where you are, or where ChatGPT thinks you are. --- ## What This Tool Helps With We built this feature for SEO and e-commerce professionals who want to understand how products surface in AI responses. The extension captures what ChatGPT receives, so you can: - **See the raw data**, not just what's displayed but the underlying payload - **Track carousel positions** and know where products rank for specific queries - **Identify metadata patterns** like cluster IDs, merchant tags, provider sources - **Export for analysis**, copy JSON, compare across queries --- ## A Note on What We Don't Know There's a lot we haven't figured out: - **Why does ChatGPT reorder products?** We see that it does, but not the logic behind it. - **What are p1 and p3?** These appear in some metadata fields but not as primary providers. - **How stable are these patterns?** Our observations come from a small set of queries in one market. We'll keep exploring. If you run queries and notice something interesting, we'd love to hear about it. ### Try It Yourself The Product Recommendations capture is live in the latest version of our extension. Run a shopping query on ChatGPT, and you'll see the data appear in your report. [Get the Extension](/try-query-fanouts-extension) --- ## What's Next This is Part 2 of our series on what happens inside ChatGPT when you ask a question. We've covered Query Fanouts and now the Shopping Connector. There's more to explore, like how citations are structured, how news results differ from web results, and how the AI decides what to include. We'll share what we find. --- ## Continue Reading **[Query Fanouts: See What Answer Engines Look For](/blogs/query-fanouts)** Part 1 - How ChatGPT breaks your question into multiple searches # Inside Google AI Mode Ads: How Shopping Ads Appear in AI Mode and What the DOM Reveals > We intercepted Google AI Mode's network logs to see how ads are served inside AI-generated responses. Full breakdown of the ad DOM structure, click tracking, product cards, and how it connects to existing Google Ads campaigns. **Published:** 2026-04-02 **Category:** Technical **URL:** https://www.tryrankly.com/blogs/google-ai-mode-ad-infrastructure-breakdown ## In this article 1. What We Did 2. No Separate Ad System 3. The Queries 4. Organic Cards vs Ads 5. Ad DOM Structure 6. Click Tracking 7. Ad Disclosure 8. ChatGPT vs Google 9. How to Prepare 10. What This Means --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/google-ai-mode-ad-infrastructure-breakdown](https://www.tryrankly.com/blogs/google-ai-mode-ad-infrastructure-breakdown). **Keywords:** Google AI Mode ads, AI Mode shopping ads, Google Ads AI Mode, aclk, gclid, Direct Offers # Universal Cart, AP2, Gemini Spark: inside Google I/O 2026 > Every Google I/O 2026 and Marketing Live 2026 announcement that mattered for agentic commerce and AI search, with the keynote video embedded. Universal Cart (cross-merchant agentic shopping cart with Nike, Sephora, Target, Walmart, Wayfair, Ulta, Shopify), UCP expansion to 20+ partners, Marketing Live UCP Tech Council expansion + Stripe-Google partnership, AP2 v0.2.0 donated to FIDO, Gemini Spark, AI Mode + Information Agents on Gemini 3.5 Flash hitting 1B MAUs, Project Mariner folded into Gemini Agent + Chrome auto-browse, Ask YouTube, Custom Generative UI, Agentic Booking expanding into local services. **Published:** 2026-05-20 **Category:** Analysis **URL:** https://www.tryrankly.com/blogs/google-io-2026-agentic-commerce-ai-search ## In this article 1. Universal Cart (cross-merchant agentic cart) 2. UCP expansion to 20+ partners and new categories 3. Marketing Live: Tech Council + Stripe-Google partnership 4. AP2 v0.2.0 donated to FIDO Alliance 5. Gemini Spark (24/7 personal agent on AI Ultra) 6. AI Mode + Information Agents (Gemini 3.5 Flash, 1B MAUs) 7. Project Mariner folded into Gemini Agent + Chrome auto-browse 8. AI Mode refresh, Ask YouTube, Custom Generative UI 9. Live Docs, SynthID, Agentic Booking expansion 10. What it means --- Read the full article with all source citations and the keynote video at [https://www.tryrankly.com/blogs/google-io-2026-agentic-commerce-ai-search](https://www.tryrankly.com/blogs/google-io-2026-agentic-commerce-ai-search). **Keywords:** Google I/O 2026, Marketing Live 2026, Universal Cart, UCP Tech Council, Stripe-Google UCP, AP2, FIDO, Gemini Spark, Information Agents, Project Mariner, Ask YouTube, Custom Generative UI # How ChatGPT Search Works: The 7-Stage Pipeline Behind Every Answer | Rankly **Page URL:** https://www.tryrankly.com/blogs/how-chatgpt-search-works **Published:** March 11, 2026 **Category:** Technical **Read Time:** 12 min --- A technical breakdown of how ChatGPT Search processes queries through 7 stages: classification, query generation, filtering, page fetch, embedding, deep-read selection, and answer generation. # How Microsoft Copilot Search Works: The Architecture Deepdive > A technical breakdown of how Microsoft Copilot processes your query through the Prometheus model, Bing grounding, 5-stage orchestration pipeline, RAG system, and Copilot Checkout shopping pipeline. **Published:** 2026-03-21 **Category:** Technical **URL:** https://www.tryrankly.com/blogs/how-microsoft-copilot-search-works ## In this article 1. Overview 2. The Three Pillars 3. Prometheus Model 4. Query Processing 5. Orchestration Pipeline 6. Grounding 7. RAG Pipeline 8. Semantic Index 9. Consumer vs Enterprise 10. Shopping 11. Citations 12. Responsible AI 13. What This Means 14. Sources --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/how-microsoft-copilot-search-works](https://www.tryrankly.com/blogs/how-microsoft-copilot-search-works). **Keywords:** Microsoft Copilot, Copilot Search, Prometheus model, Bing grounding, Microsoft Graph, semantic index, RAG, enterprise AI search, orchestration pipeline # Inside ChatGPT's New Shopping Landing Page > We ran 200 shopping-intent queries and analyzed every XHR request, every product entity, every API endpoint, and every model decision. The complete technical breakdown of ChatGPT's new shopping page. **Published:** 2026-04-10 **Category:** Technical **URL:** https://www.tryrankly.com/blogs/inside-chatgpt-shopping-page ## In this article 1. What We Did 2. The Landing Page 3. The Core Difference 4. Three New Endpoints 5. Model Routing 6. 15 Feed Partners 7. Product Schema 8. Query Fanouts 9. Image Pipeline 10. Live Ads Captured 11. What This Means --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/inside-chatgpt-shopping-page](https://www.tryrankly.com/blogs/inside-chatgpt-shopping-page). **Keywords:** ChatGPT shopping, product discovery, ACP, Sonic shopping, openai_etsy # Inside Generative Engines: A Mathematical and System-Level Breakdown **URL:** https://www.tryrankly.com/blogs/inside-generative-engines **Category:** Technical | **Read Time:** 12 min | **Date:** 25 October 2024 **Description:** A deep dive into the mathematical foundations and system architecture that power modern generative AI engines like ChatGPT, Perplexity, and Gemini. --- ## 1. The Generative Engine as a Function At its core, a generative engine is a mapping from a user query to a response: ``` f_GE: (q_u, P_U) → r ``` where **q_u** is the user's query, **P_U** is the personalization context (such as location or intent history), and **r** is the generated response (structured text with inline citations). Unlike a classical search engine that ranks documents, a GE synthesizes an answer by reading, reasoning, and rewriting through multiple neural modules. --- ## 2. The Multi-Model Pipeline A modern GE is a composition of specialized subsystems: ``` f_GE = G_resp ∘ G_sum ∘ SE ∘ G_qr ``` ### 2.1 Query Reformulation (G_qr) Expands **q_u** into semantically diverse sub-queries: ``` Q_1 = {q_1, q_2, …, q_n} ~ p(Q_1 | q_u; θ_qr) ``` ### 2.2 Retrieval Engine (SE) Fetches a ranked set of sources using information retrieval: ``` S = {s_1, s_2, …, s_m} ~ p(S | Q_1; θ_ret) ``` ### 2.3 Summarization Model (G_sum) Compresses each document into a short, citation-ready summary: ``` Sum_j = G_sum(s_j), α_j = |Sum_j| / |s_j| ``` ### 2.4 Response Synthesizer (G_resp) Constructs the final response: ``` r = G_resp(q_u, Sum) ``` --- ## 3. Sentence-Level Structure and Citations Let the response be a sequence of **o** sentences: ``` r = ⟨ℓ_1, ℓ_2, …, ℓ_o⟩ ``` Each sentence **ℓ_t** is annotated with a citation set **C_t ⊆ S**. For attribution integrity: - **Citation precision** is the fraction of citations that truly support ℓ_t. - **Citation recall** is the fraction of factual claims in ℓ_t that are cited. --- ## 4. Quantifying Visibility Inside a Generative Response Visibility in a generative engine is embedded within the synthesized text. ### 4.1 Word-Share Impression ``` Imp_wc(c_i, r) = (Σ_{s∈S_ci} |s|) / (Σ_{s∈S_r} |s|) ``` ### 4.2 Position-Weighted Impression ``` Imp_pwc(c_i, r) = (Σ_{s∈S_ci} |s| · e^{-pos(s)/|S|}) / (Σ_{s∈S_r} |s|) ``` ### 4.3 Subjective Impression ``` Subj(c_i) = [Rel, Inf, Uniq, Pos, Click, Div] Imp_subj(c_i) = Σ_k w_k · Subj_k(c_i) ``` --- ## 5. Optimization Objectives The generative engine optimizes for expected answer quality: ``` max_r E[f(Imp(c_i, r), Rel(c_i, q, r))] ``` Content creators, on the other hand, optimize visibility: ``` max_ci Imp(c_i, r) ``` This dual optimization forms the basis of Generative Engine Optimization (GEO). --- ## 6. Measuring Visibility Change After a content update, visibility improvement is defined as: ``` Improve_si = (Imp_si(r') - Imp_si(r)) / Imp_si(r) × 100 ``` Empirically, factual enrichment and structural clarity yield the highest lifts. --- ## 7. Probabilistic Model of Answer Generation Each GE stage is a stochastic mapping: ``` Q_1 ~ p(Q_1 | q_u; θ_qr) S ~ p(S | Q_1; θ_ret) Sum ~ p(Sum | S; θ_sum) r ~ p(r | q_u, Sum; θ_resp) ``` The overall likelihood of producing **r** given **q_u** is: ``` p(r | q_u) = Σ_{Q1,S,Sum} p(Q_1 | q_u) p(S | Q_1) p(Sum | S) p(r | q_u, Sum) ``` --- ## 8. DAG Representation ``` q_u → Q_1 → S → Sum → r ``` Each node performs a transformation, and each edge defines a conditional probability distribution. Critical hyperparameters include fan-out size **n**, retrieval depth **k**, summarization ratio **α**, answer length **L**, and citation density **d_c**. --- ## 9. Why GEO Works Optimized content affects two conditional probabilities: ``` Imp(s_i, r) ∝ P(s_i ∈ S | q_u) × P(s_i ∈ C_t | s_i ∈ S) ``` By improving both retrieval likelihood and synthesis attribution, GEO enables even lower-ranked sources to capture higher visibility in final LLM answers. --- ## 10. Multi-Turn Extension For conversational engines, context history **H = ⟨(q_t, r_t)⟩** conditions the next response: ``` r_{T+1} ~ p(r_{T+1} | H; θ) ``` This defines a temporal generative process that continuously updates latent context distributions. --- ## 11. Computational Characteristics If **k** is the number of retrieved sources and **L** is the total token length, then inference cost is approximately: ``` O(k · L_sum + L^2) ``` Summarization is linear in document size, while synthesis scales quadratically with attention, explaining why most engines restrict **k ≤ 5** and compress summaries aggressively. --- ## 12. Summary | Component | Role | Form | |---|---|---| | Query Reformulation | Expand queries | Q_1 ~ p(Q_1 \| q_u) | | Retrieval | Fetch sources | S ~ p(S \| Q_1) | | Summarization | Compress docs | Sum ~ p(Sum \| S) | | Synthesis | Generate response | r ~ p(r \| q_u, Sum) | | Impression | Measure visibility | Imp_wc, Imp_pwc, Imp_subj | | Optimization | GE utility | max_r E[f(Imp, Rel)] | Generative engines are probabilistic pipelines that optimize for contextual answer quality under strict latency and memory constraints. Understanding their mathematical structure is essential to improving your brand's visibility within AI-driven ecosystems. Want to know more about how Rankly is built to solve your visibility-to-conversion funnel? [Schedule a Demo](https://cal.com/rankly/30min) # Inside WebMCP: how Google's new browser API lets websites talk to AI agents natively **Date:** 2026-05-26 **URL:** https://www.tryrankly.com/blogs/inside-webmcp-google-agentic-web **Section:** Deep Dive **Read time:** 14 min WebMCP is a W3C Web Machine Learning Community Group draft co-authored by Google and Microsoft. It adds one new browser API, `navigator.modelContext.registerTool(...)`, which lets a page register JavaScript "tools" that an AI agent inside the browser can call directly. Behind a Canary flag today; origin trial pending in Chrome 149 (currently in the Early Preview Program). ## What WebMCP is A single new browser API surface: `navigator.modelContext.registerTool(tool, options)`. The spec ships no tools of its own. Each tool is a JavaScript function the page itself registers with four parts: a `name` (regex `[A-Za-z0-9_.\-]{1,128}`), a natural-language `description` the agent reads to decide whether to call it, a JSON Schema `inputSchema`, and an async `execute()`. Lifecycle is the spec's `AbortSignal` pattern - register on mount, unregister on unmount. Editors: Brandon Walderman (Microsoft), Khushal Sagar (Google), Dominic Farolino (Google). Spec home: https://webmachinelearning.github.io/webmcp/ ## Who it's for - human or agent? For agents only. Humans see no difference. The page declares its actions as tools on a separate channel (`navigator.modelContext`) that humans never touch. Closer in spirit to schema.org for AI than to a UI: schema.org describes content; WebMCP describes capabilities. ## Which agents specifically Only agents that act through a real browser tab. Five concrete categories see WebMCP tools: - **Native browser AI** (e.g. Gemini in Chrome sidebar) - **Browser extension** (e.g. Google's WebMCP Inspector) - **In-page widget** the site itself ships - **Headless browser stack** (Browserbase, Browser Use) - **CV-driven browser agent** if hooked up (Project Mariner, Anthropic Computer Use) Two categories do NOT see WebMCP: - **Server-side crawler** (GPTBot, Googlebot, ClaudeBot, Bingbot) - no browser in the request path - **Server-side LLM tool-caller** (ChatGPT plugins, server MCP) - use MCP / ACP / UCP server protocols instead ## Why it's locked behind EPP right now As of May 2026, the API is in Chrome Canary 146+ behind a `chrome://flag`. The public origin trial is announced for Chrome 149 but the registration card is not yet on the Origin Trials dashboard. Path today: Early Preview Program at https://goo.gle/chrome-ai-dev-preview-join Without an origin trial token, stable-Chrome users won't have `navigator.modelContext` exposed automatically. Three things still work for them: in-page assistants the site itself runs, Google's WebMCP Inspector extension (uses `navigator.modelContextTesting` via extension permissions), Canary users with the flag. ## Who made it and why W3C Web Machine Learning Community Group draft (not yet on the Recommendation track). Intent to Experiment filed in April 2026; publicly announced at Google I/O 2026. The gap: browser-side AI agents today have two bad options for taking action on a site: 1. **Parse the DOM and click pixels** - slow, fragile, breaks every redesign 2. **Call the site's server API directly** - requires the agent to discover, authenticate, and authorize against an auth system the user is already signed in to in the browser WebMCP's answer: the page is already running, already logged in, already has working code paths for every UI action. The agent should call those functions with structured arguments. Eliminates DOM scraping and auth-juggling. Keeps the user's tab as the consent boundary. ## How you allow / disallow specific agents through WebMCP Two origin-based controls (not User-Agent-based - WebMCP does not look at UA strings). **1. Permissions-Policy: tools header.** Standard HTTP header. Default is `self` (only your own origin can registerTool). Examples: Permissions-Policy: tools=self # only own origin Permissions-Policy: tools=(self "https://partner.example.com") # allow a partner frame Permissions-Policy: tools=() # disable WebMCP entirely **2. Per-tool options.exposedTo allowlist.** A list of agent origins allowed to see and call this specific tool. Other agents see the tool as not-existing. navigator.modelContext.registerTool( { name: 'complete_checkout', ... }, { signal: ctrl.signal, exposedTo: [ 'https://gemini-in-chrome.google', 'https://inspector.webmcp.dev', ], } ) These compose. Permissions-Policy is coarse on/off at the origin level; exposedTo is fine-grained per-tool. Add `client.requestUserInteraction(...)` on destructive tools and you have three independent gates: server-decided origin allowlist, page-decided per-tool allowlist, user-decided confirm-on-call. No User-Agent allowlist exists in the current draft. Origin-only. ## Where it sits in the stack WebMCP is a client-side mechanism. Configuration touches multiple layers: - **CDN / edge** - decides which clients fetch your HTML. Blocks here happen before WebMCP exists. - **Origin server** - sends `Permissions-Policy: tools`, ships page JS, injects the `Origin-Trial` token meta tag. - **Page JS** - calls `registerTool` with per-tool `exposedTo` and lifecycle `AbortSignal`. - **Browser** - owns `navigator.modelContext`, enforces policy, surfaces tools to in-browser agents. WebMCP is NOT a substitute for the edge. It is a different control surface for a different audience. ## Scenario A: CDN blocks everything. Can WebMCP let some agents through? No. If the edge refuses the request, the agent never gets the HTML, never sees the JS, never reaches `registerTool`. WebMCP runs three layers up. Mostly a moot scenario - Gemini-in-Chrome, the Inspector extension, and in-page widgets all run inside a normal user-opened tab; the HTTP request looks like any human visit and the CDN has nothing to block. The case where it matters: headless browser stacks like Browserbase / Browser Use that hit your CDN with identifiable User-Agents or IP ranges. If your edge blocks them, WebMCP doesn't change anything. Outcome matrix: | Edge says | Page WebMCP config | Result | |---|---|---| | Blocked | Allowed (any) | Edge wins. 403. No WebMCP. | | Allowed | `tools=()` Permissions-Policy | Page loads, registry denied, zero tools. | | Allowed | `tools=self`, no exposedTo | Every browser-side agent sees every tool. | | Allowed | `tools=self` + per-tool `exposedTo:[A,B]` | Only origins A and B see gated tools. | ## Scenario B: CDN allows everyone. Can WebMCP selectively block some? Yes - with one asterisk. - Send `Permissions-Policy: tools=()` to disable WebMCP entirely on the response. Page renders normally for humans; no agent can call the registry. - Keep WebMCP enabled at origin level, use per-tool `exposedTo` to limit tools to specific agent origins. Everyone else loads the page fine. **Asterisk:** WebMCP gates by *origin of the calling agent*, not by "is this a bot." You cannot say "allow OpenAI agents only" - you can say "allow this specific browser-side surface only." Server-side bots are unaffected because they never run in a browser. ## How agents actually arrive on a WebMCP-enabled page Three concrete vectors today: 1. **Native browser AI (e.g. Gemini in Chrome).** Built into the browser. Reads the tool registry via `navigator.modelContext`. The "every Chrome user gets an agent for free" story. Gated on the Chrome 149 origin trial. 2. **Browser extension (Google's WebMCP Inspector).** Installed once, works on any site. Uses `navigator.modelContextTesting` via extension permissions, so no origin trial token needed. Lists tools, lets you execute manually, runs a BYO-key chat (Gemini, OpenAI, Anthropic, Ollama). 3. **In-page widget shipped by the site.** A "chat with us" widget that's actually an agent reading the page's own tool registry. Works for every visitor with no install. Browserbase / Browser Use style headless stacks fit the same shape. None of these arrive through a request the CDN can identify as "an agent." The HTTP request looks like a normal user's browser, *because it is one*. The agent is acting *through* the browser, not as the browser. ## Our demo: WebMCP on the Rankly Merch Store Forked the live store, layered WebMCP on top. Shipped: - **16 globally-registered tools** on every page: search_products, get_product, list_categories, view_cart, add_to_cart, navigate, best_sellers, find_deals, sign_in, sign_in_demo, register_user, authenticate, local_availability, track_order, returns_policy, faq_support. - **Page-scoped tools** registered only where they make sense: `select_variant` / `set_quantity` / `find_in_store` / `leave_review` on product detail, cart-mutators on cart, address + checkout-completer on checkout. Lifecycle uses AbortSignal. - **Two ways to drive the registry from the page itself.** A demo mode (hardcoded intent router, no API key, 181/181 regression cases) and a live mode (Claude Messages API direct from browser, chips protocol). - **Two external surfaces** the same registry serves automatically: Google's WebMCP Inspector (auto-detects all 16 globals via `navigator.modelContextTesting`), and any Canary user with the WebMCP flag. - **Zero backend changes.** WebMCP is client-side. The store's UCP / ACP / server MCP endpoints are untouched. Tested side by side with the Inspector. The agent can search, sign in, add to cart, check out, place an order - on the user's real authenticated session, without parsing the DOM. ## What it means for merchants WebMCP does not replace server-side protocols. ACP for ChatGPT shopping, UCP for Google Universal Cart, server MCP for LLM-host integrations - those are different surfaces for different agents (server-side, not browser-side). WebMCP adds a fourth surface: browser-side agents acting on behalf of your logged-in user, in their own tab. When Gemini-in-Chrome rolls out broadly and the origin trial opens to stable Chrome, every WebMCP-registered site becomes drivable from a sidebar 65% of internet users already have. Sites that haven't registered their actions fall back to DOM scraping with all the latency and reliability that entails. The thing to watch is the Chrome 149 origin trial. The day the dashboard listing opens, registering your origin is a 10-minute job that converts "works for Canary devs and extension users" into "works for everyone on stable Chrome." ## References - WebMCP spec: https://webmachinelearning.github.io/webmcp/ - Chrome for Developers - WebMCP: https://developer.chrome.com/docs/ai/webmcp - WebMCP EPP announcement: https://developer.chrome.com/blog/webmcp-epp - Intent to Experiment: https://groups.google.com/a/chromium.org/g/blink-dev/c/gmYffo5WOE8/m/OJxuQRP3AAAJ - WebMCP Inspector (Chrome Web Store): https://chromewebstore.google.com/detail/webmcp-model-context-tool/gbpdfapgefenggkahomfgkhfehlcenpd - Inspector source: https://github.com/beaufortfrancois/model-context-tool-inspector - Google I/O 2026 recap: https://www.tryrankly.com/blogs/google-io-2026-agentic-commerce-ai-search # Introducing Rankly Agent Directory > A free public catalog of 4,171 AI bots, crawlers, and agents hitting websites today. What each one wants, how to identify it, and whether to allow or block it. **Published:** 2026-05-11 **Category:** Product **URL:** https://www.tryrankly.com/blogs/introducing-agent-directory ## In this article 1. Why we built the Agent Directory 2. What the directory shows for each bot 3. How to filter and find what matters 4. Per-bot detail pages: identification, control, expected behavior 5. Operator roll-ups: every bot from one vendor in a single table 6. Data sources (BuzzStream, Cloudflare Radar AI Insights, Vercel) 7. Lead-in to Agent Analytics + Agent Experience --- Read the full article with screenshots and the live directory at [https://www.tryrankly.com/blogs/introducing-agent-directory](https://www.tryrankly.com/blogs/introducing-agent-directory). **Keywords:** Agent Directory, AI bot directory, AI crawler list, GPTBot, ClaudeBot, PerplexityBot, robots.txt, block AI crawlers, AEO, agent analytics # Introducing Rankly Protocol Tracker > A live feed of merged pull requests, releases, and spec changes across the open-source protocols shaping agentic commerce. **Published:** 2026-05-18 **Category:** Product **URL:** https://www.tryrankly.com/blogs/introducing-protocol-tracker ## In this article 1. Why this is required 2. The protocols we watch, grouped into Commerce, Payments, Connectors, Discovery --- Read the full article with screenshots and the live tracker at [https://www.tryrankly.com/blogs/introducing-protocol-tracker](https://www.tryrankly.com/blogs/introducing-protocol-tracker). **Keywords:** Protocol Tracker, agentic commerce, ACP, UCP, AP2, MCP, A2A, x402, MPP, NLWeb, WebMCP, llms.txt, Web Bot Auth, L402, Visa TAP, Verifiable Intent # Now You Can Personalize Answers Inside ChatGPT > How a brand can show up inside ChatGPT with branded product cards, nearby-store stock, and native checkout. The four personalization layers - brand, context, intent, memory - and what an Apps SDK / MCP storefront unlocks beyond AEO. **Published:** 2026-05-01 **Category:** Analysis **URL:** https://www.tryrankly.com/blogs/personalize-answers-inside-chatgpt ## In this article 1. The Shift 2. Answers vs Apps: Two Surfaces, Two Ceilings 3. What an App Can Render 4. What Signals Actually Exist Inside a Chat 5. The Four Layers of Personalization (Brand, Context, Intent, Memory) 6. An Omnichannel Walkthrough: The Linen T-Shirt 7. See It In Action 8. The Feedback Loop 9. Outlook --- Read the full article with all source citations and the in-chat demo at [https://www.tryrankly.com/blogs/personalize-answers-inside-chatgpt](https://www.tryrankly.com/blogs/personalize-answers-inside-chatgpt). **Keywords:** ChatGPT, OpenAI Apps SDK, MCP, agentic commerce, personalization, in-chat checkout, AEO, brand surface # Query Fanouts: See What Answer Engines Look For **URL:** https://www.tryrankly.com/blogs/query-fanouts **Category:** Product | **Read Time:** 5 min | **Date:** 26 December 2024 Discover how LLMs break down your queries into multiple search queries called query fanouts, and learn how to use them to improve your AI visibility. --- ## What Are Query Fanouts? When a large language model (LLM) performs a web search on behalf of a user, it rarely runs just one query. Instead, it breaks the original prompt into multiple smaller, related search queries. These sub-queries are called **query fanouts**, and they reveal what the model is actually searching for behind the scenes. > Think of query fanouts as the hidden roadmap an AI uses to research a question before generating its final answer. --- ## How Do Query Fanouts Work? Query fanouts are generated whenever an LLM answers a question with web search enabled. The model refines the intent of the original prompt and sends out a set of high-intent search queries that help it build the final response. ### Where to See Query Fanouts You can view them in several ways: - **API responses:** When web search is enabled, the API includes the fanout queries in the network call results. - **Network developer tools:** In the browser, using the session/message ID you can inspect the Network tab to see the fanouts. --- ## Query Fanouts by Platform Different AI platforms implement query fanouts in their own ways. Here's how the major players handle them: ### Query Fanouts in ChatGPT ChatGPT uses query fanouts when its web browsing feature is enabled. The model breaks down complex queries into targeted sub-searches to gather comprehensive information before synthesizing a response. | Metadata | Description | |----------|-------------| | title | Auto-generated conversation title | | conversation_id | Unique ID for the chat session | | search_queries | The search terms ChatGPT uses to find information | | citations | Links to websites used in the response | | search_turns_count | Number of web searches performed | ### Query Fanouts in Perplexity Perplexity is built specifically for AI-powered search and makes extensive use of query fanouts to provide comprehensive, well-sourced answers. | Metadata | Description | |----------|-------------| | query | The original search question | | web_results | List of websites searched and used | | citations | Numbered references to sources in the answer | | domain | Source website names | ### Query Fanouts in Claude Claude by Anthropic also utilizes query fanouts when connected to web search tools, decomposing complex queries into focused searches. | Metadata | Description | |----------|-------------| | name | Conversation title | | web_search.query | The search terms Claude uses to find information | | knowledge_results | List of websites found during search | | citations | Links referenced in the final response | --- ## How Does This Help Your Brand? If your brand is not mentioned in AI-generated answers, it means you likely don't have content that directly addresses the query fanouts the model is using. To get mentioned, you need to create pages that **directly target those specific query fanouts**. > Understanding query fanouts transforms content strategy from guesswork into data-driven optimization. --- ## How Rankly's Query Fanout Extension Works Our Query Fanouts Extension makes it easy to discover exactly what AI models are searching for: 1. **Enter Your Query** - Type in any search query you want to analyze. 2. **See the Fanouts** - View the complete list of sub-queries the AI generates. 3. **Optimize Your Content** - Create targeted content for each fanout to improve your AI visibility. --- ## How We Use Query Fanouts Internally We use query fanouts as the foundation of our content and optimization strategy. For every tracked query, we show: - **The complete list of fanouts:** Every sub-query the AI model generates. - **Coverage analysis:** Which fanouts your site already covers. - **Actionable recommendations:** Specific suggestions for missing pages and content gaps. This removes guesswork from content planning. Instead of optimizing for a single keyword, you optimize for the **full set of queries** an AI model actually uses. --- ## Get Started Curious about what AI models are actually searching for when users ask about your brand or industry? **Try our Query Fanouts Extension to discover exactly what AI searches for your queries—and much more!** ### Try the Query Fanouts Extension Uncover the hidden queries AI uses to answer questions about your brand. [Get Started](/try-query-fanouts-extension) # Shopify's Recent Announcements on Agentic Commerce > A breakdown of Shopify President Harley Finkelstein's Upfront Summit 2026 talk on agentic storefronts, UCP, the hub-and-spoke model, OpenAI's Instant Checkout reversal, and what merchants need to do now. **Published:** 2026-03-24 **Category:** Insights **URL:** https://www.tryrankly.com/blogs/shopify-agentic-commerce-announcements ## In this article 1. The New Front Door 2. Hub-and-Spoke Model 3. Why Agents Change Discovery 4. Universal Commerce Protocol 5. The Fashion Nova Story 6. OpenAI Killed Checkout 7. Agentic Storefronts Are Live 8. Shop Pay as Trust Layer 9. 5 Merchant Imperatives --- Read the full article with all source citations and analysis at [https://www.tryrankly.com/blogs/shopify-agentic-commerce-announcements](https://www.tryrankly.com/blogs/shopify-agentic-commerce-announcements). **Keywords:** Shopify, Agentic Commerce, Harley Finkelstein, UCP, Agentic Storefronts, OpenAI, Shop Pay, Universal Commerce Protocol # Stripe's Recent Announcements on Agentic Commerce > Stripe shipped 288 products at Sessions 2026 and framed the entire keynote around agents as autonomous economic actors. These are the agentic commerce announcements worth tracking. **Published:** 2026-05-05 **Category:** Analysis **URL:** https://www.tryrankly.com/blogs/stripe-agentic-commerce-announcements ## In this article 1. Five levels of agentic commerce 2. Buy inside the four AI surfaces 3. Agentic Commerce Suite 4. Checkout Studio 5. Order Intents and Machine Payments Protocol 6. Link Wallet for Agents 7. Agents buying from agents 8. Tempo and streaming stablecoins 9. Radar for AI businesses 10. Stripe Console 11. Developer surface for agents 12. Agent guardrails --- Read the full article with all source citations at [https://www.tryrankly.com/blogs/stripe-agentic-commerce-announcements](https://www.tryrankly.com/blogs/stripe-agentic-commerce-announcements). **Keywords:** Stripe Sessions 2026, agentic commerce, Stripe Link, Order Intents, Machine Payments Protocol, Tempo, Radar AI, Stripe Console, Checkout Studio # UCP Tech Council Expansion: Recent Updates and Full Architecture Breakdown > Amazon, Meta, Microsoft, Salesforce, and Stripe joined the UCP Tech Council on 24 April 2026. Full architecture breakdown of the Universal Commerce Protocol: the six capabilities, the /.well-known/ucp manifest, the Trust Triangle, the partner roster, the timeline, and how UCP relates to A2A, AP2, MCP, and ACP. **Published:** 2026-04-26 **Category:** Technical **URL:** https://www.tryrankly.com/blogs/ucp-tech-council-expansion-recent ## In this article 1. What is UCP 2. Official Resources 3. Who Developed It 4. Partners 5. Who It's For 6. UCP vs A2A, AP2, MCP 7. Architecture 8. Discovery Manifest 9. The Six Capabilities 10. The Trust Triangle 11. Transports 12. Code and SDKs 13. Timeline 14. UCP vs ACP 15. Outlook --- Read the full article with all source citations, illustrations, and analysis at [https://www.tryrankly.com/blogs/ucp-tech-council-expansion-recent](https://www.tryrankly.com/blogs/ucp-tech-council-expansion-recent). **Keywords:** Universal Commerce Protocol, UCP, agentic commerce, Google, Shopify, AP2, A2A, MCP, ECP, ACP, Agentic Commerce Protocol, Stripe, OpenAI, Trust Triangle, UCP Tech Council, AI shopping protocol # What is AEO? Why Generative Engine Optimization is the New SEO **URL:** https://www.tryrankly.com/blogs/what-is-aeo **Category:** Insights | **Read Time:** 6 min | **Date:** 5 November 2024 Understand how Answer Engine Optimization is reshaping digital visibility. From citations to sentiment, discover why ranking in AI Answer Engines matters more than ever. --- ## Introduction Search has always been the gateway to the web. For two decades, Search Engine Optimization (SEO) was the discipline of winning Google's "ten blue links." But today, search is being radically transformed by Answer Engines such as ChatGPT, Perplexity, Claude, Gemini, and Google's AI Overviews. Unlike Google's traditional SERPs, these engines do not just rank pages. They generate answers, weaving information from multiple sources into a single authoritative response. This seismic shift means SEO alone is no longer enough. What matters now is whether your brand is cited inside the AI-generated answer. This is the realm of Answer Engine Optimization (AEO), the next frontier of search. --- ## What is Answer Engine Optimization (AEO)? Answer Engine Optimization is the practice of optimizing your content for AI-driven answer engines, ensuring your brand is cited, visible, and accurately represented in generative search. Instead of measuring success in rankings and clicks, AEO focuses on: - **Citations:** Does the AI quote your brand or page? - **Visibility:** How often does your brand appear in generated answers? - **Impression Metrics:** How deeply, prominently, and positively are you represented? - **Sentiment and Share of Voice:** Do AIs frame you as authoritative or secondary? > Put simply: SEO optimizes for human clicks, AEO optimizes for AI citations. --- ## Why Traditional SEO is No Longer Enough The data is clear: AI-driven search changes user behavior. - In Google's Search Generative Experience (SGE), link clicks drop by nearly 50% when an AI summary is present, from 15% without a summary to 8% with one. - Roughly 26% of AI-assisted searches end with zero clicks. Users get their answer from the AI and move on. - Perplexity.ai alone handled 780 million queries in May 2025, proof that AI-first discovery is no longer niche. This means visibility now depends on being inside the answer itself. > As SEO veteran Rand Fishkin noted: "AI results will increasingly favor large brands, making it harder for smaller players to compete unless they establish undeniable niche authority." --- ## The Rise of Generative Engine Optimization (GEO) Researchers at Stanford and Georgia Tech have coined the term Generative Engine Optimization (GEO), the science of optimizing for AI-driven discovery. More recently, comprehensive research from the University of Toronto has provided empirical analysis of how AI search engines systematically favor earned media over brand-owned content. Unlike SEO, which relied on backlinks and keyword density, GEO emphasizes: - Language signals that LLMs recognize as authoritative. - Structured data and clarity for easy parsing. - Citations, statistics, and quotations that AI prefers to incorporate. - Impression metrics such as visibility score, depth of mention, and diversity of references. > Andreessen Horowitz summarized it best: "Traditional search was built on links; generative search is built on language." --- ## SEO vs AEO: Key Differences | SEO | AEO | |-----|-----| | Optimizes for human rankings | Optimizes for AI-generated citations | | Focuses on keywords, backlinks, CTR | Focuses on visibility, impression metrics, sentiment | | Competes for top 10 blue links | Competes for inclusion in one synthesized answer | | Success = traffic and clicks | Success = citations, influence, representation | | Tools: Google Analytics, Ahrefs, SEMrush | Tools: Citation tracking, Visibility Scores, AEO dashboards | --- ## Why AEO Matters for Brands - **Zero-click future.** If you are not cited, you are invisible. - **Zero-sum competition.** If a competitor earns the citation, you lose it. - **Representation risk.** AI can misframe your brand unless you track sentiment and influence. - **Hybrid reality.** SEO and AEO must co-exist. Google still drives traffic, but AI drives perception. According to McKinsey, brands that adapt early to generative AI in marketing can expect a 15–20% lift in digital visibility compared to late adopters. --- ## How to Get Started with AEO ### 1. Audit Your Visibility Run tests across AI engines. Are you cited? How often? What is the sentiment? ### 2. Benchmark Competitors Who gets mentioned more? Do competitors dominate earned media citations? ### 3. Optimize for Clarity and Authority Use structured headings, tables, and concise facts. AI models prefer scannable, verifiable information. ### 4. Add Statistics, Quotes, and Sources Research shows that adding quotations, statistics, and citations can increase AI visibility by up to 40%. > "Success in AEO will come from creating content worth citing, not chasing tricks to game the system." – Aggarwal et al., Generative Engine Optimization Framework (2024) ### 5. Invest in Earned Media AI engines heavily favor third-party, authoritative sites. Appear in industry press, reviews, and expert outlets to increase trust. ### 6. Adopt a Hybrid Strategy SEO brings traffic. AEO builds authority in generative answers. Winning brands must do both. --- ## Conclusion Search is entering its most profound transformation since Google launched. SEO shaped the last 20 years; AEO will shape the next 20. Brands that adapt early will control their narrative inside AI answers, shaping customer perception before the click ever happens. Those that wait will fade into invisibility, not because their site is not indexed, but because their name is never cited. At Rankly, we believe answer engines should not remain black boxes. Answer Engine Optimization makes visibility measurable, influence explainable, and citations actionable. **The future of discovery belongs to AEO. The only question is: will your brand be part of the answer, or left out of it?** --- ## References [1] Pew Research Center. (2025). "Google users are less likely to click on links when an AI summary appears in the results." [2] Perplexity AI. (2025). "CEO says Perplexity hit 780M queries in May 2025." [3] Aggarwal, P., et al. (2024). "GEO: Generative Engine Optimization." arXiv:2311.09735. [4] Andreessen Horowitz. (2023). "The Future of Search: From Links to Language." [5] McKinsey & Company. (2024). "Generative AI's Impact on Marketing and Digital Visibility." [6] Fishkin, R. (2024). "Large Brand Bias in AI Search Results." SparkToro Insights. [7] Chen, M., et al. (2025). "Generative Engine Optimization: How to Dominate AI Search." University of Toronto. # Agent Analytics **URL:** https://www.tryrankly.com/features/agent-analytics --- ## See which agents are really on your site Most analytics were made for human visitors and never see AI agents. Rankly is built for them, measured at your edge where every agent is visible. --- ## Features ### Identify every agent that visits Count every agent at the edge, even when it never runs a tag. Thousands of bots from every major AI platform, each named and classified by what it does. ### Verify which agents are real Every request is matched to the vendor's published IP range, then marked verified or spoofed. Verify each bot against vendor IP ranges and auth signatures. ### Catch agents disguised as browsers Bots faking a real browser fall out at each check until the disguised one is flagged. Headless clients, dead browsers, fake Windows details, and very outdated Chrome are all caught. ### Know why each AI is crawling you Group bots by what they do: train, index, fetch, or shop. Every hit is split into training, search-index and live-fetch, so you see the intent. - Model training crawls - Answer-engine corpora - Real-time retrieval - Agentic commerce ### Map where agents connect from See where agents connect from across countries and networks. Each agent is plotted by origin, so a GPTBot from the wrong region stands out at a glance. ### Follow agents through the buy flow Agent requests are stitched into one session, showing the furthest step each shopper reached. --- ## Agent Directory Browse the Agent Directory: thousands of bots from every major AI platform, each named and classified by what it does. Includes GPTBot, ClaudeBot, PerplexityBot, Gemini-Deep-Research, Meta-ExternalAgent, and more. --- ## Built for AI traffic, not page views The only agent analytics that plugs straight into Shopify, alongside every major CDN provider, measuring traffic right at the source. Connect Shopify, your CDN, or your edge in minutes. --- Start measuring agents at https://www.tryrankly.com/features/agent-analytics # Agentic Commerce **URL:** https://www.tryrankly.com/features/agentic-commerce --- ## AI agents are becoming the new shoppers. Make sure your products get recommended when they browse, compare, and buy A middleware layer that sits on your existing Shopify, WooCommerce, BigCommerce store and exposes it to AI agents via UCP and MCP. --- ## Key Benefits ### AI Agent Tracking Monitor when AI agents recommend your products across all major platforms. ### Competitive Intel See exactly which competitors get recommended instead of you, and why. ### Win Rate Analytics Track your recommendation win rate and identify optimization opportunities. ### Instant Alerts Get notified when competitors overtake you in AI recommendations. --- ## Product ### Step 01: UCP Profile — Auto-generate your /.well-known/ucp profile. **One endpoint. Full store exposure.** UCP (Universal Commerce Protocol) is the emerging standard for how AI agents discover and interact with stores. Think of it like robots.txt, but for commerce capabilities. Rankly scans your store and generates a hosted UCP profile automatically. ### Step 02: API Adapters — We translate your platform's API into UCP. **You don't change anything.** Every ecommerce platform has its own API. Rankly's adapters sit between your platform and the UCP spec. When an AI agent calls UCP Product Search, we translate that into the right Shopify Storefront API query behind the scenes. #### UCP Endpoint to Shopify API Translation | UCP Endpoint | Shopify API | |-------------|-------------| | UCP Product Search | Storefront API · Products query | | UCP Product Details | Storefront API · Product by ID | | UCP Product Reviews | Storefront API · Metafields | | UCP Cart Operations | Storefront API · Cart mutations | | UCP Checkout Session | Storefront API · Cart + Checkout | | UCP Inventory Check | Admin API · Inventory levels | | UCP Price & Offers | Storefront API · Price rules | | UCP Shipping Rates | Admin API · Carrier service | | UCP Order Webhooks | Admin API · Order webhooks | | UCP Fulfillment Status | Admin API · Fulfillment API | | UCP Returns & Refunds | Admin API · Refund endpoints | | UCP Customer Profile | Admin API · Customer data | | UCP Wishlist Sync | Storefront API · Metafields | | UCP Store Locations | Admin API · Locations endpoint | | UCP Product Variants | Storefront API · Variants query | | UCP Discount Codes | Admin API · Price rules + codes | | UCP Collection Browse | Storefront API · Collections query | ### Step 03: MCP Server — Auto-generated MCP server for your store. **Every AI agent that supports MCP can now shop natively.** MCP (Model Context Protocol) is how AI agents like Claude and Cursor interact with external tools natively. Instead of calling a REST API, the agent has direct tool access — search_products, add_to_cart, create_checkout. Rankly generates your MCP server automatically. --- ## Platform Coverage Google AI Overviews, Gemini, ChatGPT, Claude, Shopify, BigCommerce, WooCommerce, Adobe Commerce, SAP Commerce Cloud, Salesforce Commerce Cloud, Etsy, Amazon --- ## The Shift — Commerce is being rewritten AI agents are replacing traditional search. Customers now describe what they want, and AI surfaces the best options, with direct purchasing built in. ### Traditional Search | Aspect | Value | |--------|-------| | Discovery | Keywords & links | | Experience | Search → Click → Browse | | Conversion | Manual cart & checkout | | Trust | Reviews & ratings | | Optimization | SEO & paid ads | **Warning:** Traditional product feeds fail on AI shopping platforms ### Agentic Commerce | Aspect | Value | |--------|-------| | Discovery | Intent & conversation | | Experience | Ask → Recommend → Buy | | Conversion | Chat-to-purchase | | Trust | AI curation & context | | Optimization | Agent visibility | **Note:** Each AI platform requires unique data structures for optimal performance The shift from ad-based search to **intent, context, and trust** is happening now. --- ## The Rise of Agentic Commerce ### 01 — ChatGPT Instant Checkout (Sep 2025) **OpenAI** OpenAI launches Instant Checkout with Shopify (1M+ merchants) and Etsy. The open-source Agentic Commerce Protocol is co-developed with Stripe for secure payments. - Key Players: Shopify, Etsy - [Read More](https://openai.com/index/buy-it-in-chatgpt/) ### 02 — Universal Commerce Protocol (Jan 2026) **Google × Shopify** Google and Shopify co-develop UCP, an open standard for agentic commerce endorsed by 20+ retailers including Target, Walmart, Best Buy, and payment processors Stripe, Visa, Mastercard. - Key Players: Google, Shopify - [Read More](https://www.shopify.com/news/ai-commerce-at-scale) ### 03 — Google AI Mode Shopping (Jan 2026) **Google** Direct shopping integration launches in Google Search AI Mode and Gemini. New Business Agent tool enables retailers to participate in AI-powered commerce at scale. - Key Players: Google, Gemini - [Read More](https://blog.google/products/ads-commerce/agentic-commerce-ai-tools-protocol-retailers-platforms/) ### 04 — Claude: No Ads, Only Commerce (Feb 2026) — Latest **Anthropic** Anthropic announces Claude will remain ad-free forever, instead focusing on merit-based agentic commerce where products are recommended based on quality, not ad spend. - Key Players: Claude - [Read More](https://www.anthropic.com/news/claude-is-a-space-to-think) --- ## Platforms Tracked — Monitor every AI shopping agent We track shopping recommendations across all major AI platforms where consumers ask agents to help them buy. | Platform | Shopping Queries | |----------|-----------------| | ChatGPT | 45M+ shopping queries/month | | Perplexity | 12M+ shopping queries/month | | Claude | 8M+ shopping queries/month | | Gemini | 25M+ shopping queries/month | | Copilot | 15M+ shopping queries/month | --- ## How It Works — From invisible to recommended ### 01 — Connect your product catalog Sync your product data via API, CSV, or direct integration with your e-commerce platform. ### 02 — We monitor AI shopping agents Our system tracks shopping queries across ChatGPT, Perplexity, Claude, Gemini and more. 24/7. ### 03 — See your recommendation status Know exactly when, where, and why AI agents recommend your products (or your competitors). ### 04 — Optimize and win Get specific recommendations to improve your product data and increase your AI recommendation rate. --- ## Success Stories > "We had no idea AI agents were recommending our competitors until Rankly showed us. Within 3 months, we went from invisible to #2 in AI shopping recommendations." > — David Chen, CMO, TechGear Direct > "The future of commerce is agentic, and Rankly helps us stay ahead. Our AI recommendation rate increased 156% after implementing their optimization suggestions." > — Sarah Martinez, Head of Digital, LuxuryBrand Co > "Finally, visibility into the black box of AI shopping. We now know exactly why ChatGPT recommends certain products and how to win those recommendations." > — James Wilson, E-commerce Director, SportsPro --- ## Get Started **Free Analysis** ### Win every AI shopping conversation [Start Free Analysis](https://cal.com/rankly/30min) # Answer Engine Analytics **URL:** https://www.tryrankly.com/features/answer-engine-analytics --- ## Understand how AI search engines discover, interpret, and recommend your content to users Track your visibility across ChatGPT, Perplexity, Gemini, and more. Monitor how AI platforms mention, rank, and recommend your brand in real time. --- ## Key Benefits ### Visibility Tracking Monitor your brand presence across all major AI platforms. ### Prompt Testing Test how AI responds to queries about your products. ### Sentiment Analysis Measure brand perception in AI-generated responses. ### Citation Monitoring Track which sources AI cites for your brand. --- ## Full-Funnel AEO — Full-funnel answer engine optimization, from discovery to conversion ### Discovery & Indexing Ensure AI engines can find, crawl, and understand your content. ### Content Optimization Structure content so AI models cite and recommend your brand. ### Performance Tracking Measure visibility, citations, and sentiment across every AI platform. ### Conversion Attribution Connect AI-driven traffic to revenue with end-to-end funnel analytics. ### AEO Funnel Metrics | Stage | Value | |-------|-------| | Visibility | 78% | | Citations | 42% | | Traffic | 2,847 | | Conversion | 3.2% | | Revenue | $12.4K | --- ## Supported AI Platforms Supported across all major AI platforms: - Google AI Overviews - ChatGPT - Perplexity - Copilot - Gemini - Claude - Meta AI - Amazon Rufus --- ## Live Dashboard — Your AI visibility at a glance Real-time metrics from all major AI platforms, unified in one powerful dashboard. ### Platform Scores | Platform | Score | Change | |----------|-------|--------| | ChatGPT | 847 | +24% | | Perplexity | 623 | +18% | | Claude | 512 | +15% | | Gemini | 445 | +12% | | Copilot | 389 | +8% | ### Metrics | Metric | Value | Change | |--------|-------|--------| | Visibility Score | 2,816 | +18.5% | | Total Citations | 1,247 | +32.1% | | Sentiment Score | 94.2% | +5.3% | ### Dashboard Stats | Stat | Value | |------|-------| | Queries Tracked | 12.4K | | Brands Monitored | 847 | | Avg Response Time | < 2s | | Data Points/Day | 1.2M | --- ## Analytics Dashboard — Data at a glance Real-time visibility metrics from ChatGPT, Perplexity, Claude, Gemini and more. All in one dashboard. ### Visibility Trends Track brand mentions across AI engines over time. ### Depth of Mention See where your brand appears in AI responses. ### Sentiment Analysis How AI describes your brand across platforms. - Positive: 67% - Neutral: 25% - Negative: 8% ### Citation Share Track owned, earned & social citations over time. ### Platform Compare your brand's performance across ChatGPT, Gemini, Claude, Perplexity, and others. #### Citation Mentions by Platform | Platform | Mentions | |----------|----------| | ChatGPT | 2.4K | | Perplexity | 1.8K | | Claude | 1.2K | ### Filter & Segment Slice your data by competitor, persona, region & more. Available filters: - Competitors - URLs - Persona - Topics - Region - Language --- ## How It Works — See what AI is really saying about you We don't rely on APIs. Our platform emulates real user behavior to capture exactly what your audience sees. ### Prompt Optimization Generate targeted prompts for every user persona, topic, and product category. Maximize your visibility coverage across all AI engines. ### Actionables on What's Working and What Needs Attention Across Owned, Earned and Social Surfaces. For every Platform, Competitor, Persona, Topics and Product Category. --- ## Query Fanouts — One question, Multiple searches When you ask an LLM a complex question, it breaks your query into multiple targeted searches. Rankly tracks every single one. ### Example **Original Query:** "What are the best noise-canceling headphones under $300 for working from home?" **Fanout Queries:** 1. best noise canceling headphones 2025 2. work from home headphones reviews 3. headphones under $300 comparison 4. Sony WH-1000XM5 vs Bose QC45 --- ## Tracking & Monitoring — Continuous visibility intelligence Set up automated tracking for your brand across AI platforms. Monitor trends, spot issues, and get alerts when visibility changes. ### Filter Parameters (8 dimensions) - **Tracking Frequency:** Daily, Weekly, Bi-weekly, Tri-weekly - **Competitors:** Sony, Bose, Apple, Samsung - **Platforms:** ChatGPT, Perplexity, Claude, Gemini - **Personas:** Tech Enthusiast, Casual User, Audiophile, Professional - **Topics:** Sound Quality, Battery Life, Comfort, Price - **Products:** WH-1000XM5, AirPods Pro, QC45, Galaxy Buds - **Regions:** United States, Europe, Asia Pacific, Global - **Languages:** English, Spanish, German, French ### Tracked Prompts | Prompt | Trend | Change | Citations | |--------|-------|--------|-----------| | Best wireless earbuds for tech enthusiasts 2025 | Up | +24% | 847 | | Sony vs Bose noise canceling comparison | Up | +18% | 623 | | Audiophile headphones under $400 reviews | Down | -8% | 412 | | Battery life comparison wireless headphones | Up | +31% | 789 | --- ## Testimonials > "Rankly completely changed how we approach AI visibility. We went from guessing to knowing exactly where we stand across every major AI platform." > — Sarah Chen, Head of Growth, TechStartup > "The citation tracking alone is worth it. We've increased our AI visibility by 340% in just 3 months by optimizing based on Rankly's insights." > — Marcus Johnson, SEO Director, MediaCorp > "Finally, a tool that takes AI search seriously. Our competitors still haven't figured out what we did. Rankly is our secret weapon." > — Elena Rodriguez, CMO, FinanceApp --- ## Start Monitoring **Free Trial** ### Win every AI shopping conversation [Book a Demo](https://cal.com/rankly/30min) # Shopping Analytics **URL:** https://www.tryrankly.com/features/shopping --- ## Track how AI recommends products and drives purchasing decisions Monitor product visibility, retailer rankings, and pricing across AI shopping agents like ChatGPT, Perplexity, and Gemini. All in one dashboard. --- ## Features ### Shopping Responses See how AI recommends products across conversational shopping queries. ### Product Visibility Track how often your products appear in AI shopping recommendations. ### Retailer Visibility Track which retailers appear in AI shopping recommendations for your products. ### Pricing & Reviews Compare avg prices, ratings, and reviews for products across AI responses. --- ## Product Visibility Dashboard How often your products appear in AI shopping recommendations. - **Visibility Score:** 72.50% (up 15%) ### Top Products by Visibility (15 tracked) | Rank | Product | Brand | Score | Trend | |------|---------|-------|-------|-------| | 1 | Anker Soundcore Space One | Anker | 89% | Up | | 2 | Soundcore Q45 ANC | Soundcore | 84% | Down | | 3 | Beats Solo 4 | Beats | 78% | Up | | 4 | JBL Tune 720BT | JBL | 72% | Stable | | 5 | Marshall Major IV | Marshall | 65% | Down | --- ## Retailer Visibility Which retailers appear in AI shopping recommendations. | Retailer | Category | Score | Products | Appearances | Trend | |----------|----------|-------|----------|-------------|-------| | Amazon | Marketplace | 94% | 48 | 312 | Up | | Best Buy | Electronics | 87% | 36 | 248 | Up | | Walmart | Marketplace | 76% | 42 | 189 | Down | | Target | Retail | 68% | 24 | 142 | Up | | B&H Photo | Electronics | 61% | 18 | 96 | Down | | Newegg | Electronics | 45% | 14 | 64 | Stable | --- ## Pricing & Reviews Avg prices, ratings, and reviews across AI responses. | Product | Avg. Price | Rating | Reviews | |---------|-----------|--------|---------| | Anker Soundcore Space One | $99.99 | 4.5 stars | 12.8k | | Beats Solo 4 | $149.95 | 4.5 stars | 24.3k | --- ## Get Started **Free Trial** ### Win every AI shopping conversation [Start 14-day Free Trial](#pricing) # Traffic Analytics **URL:** https://www.tryrankly.com/features/traffic-analytics --- ## Track and analyze traffic from AI platforms with precision attribution and conversion insights Track AI crawler visits alongside human sessions. Understand what bots are indexing, and what content converts real visitors. --- ## Key Benefits ### AI Referral Tracking Track traffic from ChatGPT, Perplexity, Claude & more. ### Conversion Attribution Compare AI-referred vs traditional search conversions. ### Journey Mapping Map the full AI discovery to purchase path. ### Real-time Dashboards Monitor AI traffic patterns as they happen. --- ## AI Agent Analytics — See how AI crawlers interact with your site Detailed insights into bot behavior, crawl patterns, and content extraction. ### Bot Identification Detect all AI crawlers by user-agent. Know exactly which bots visit your site. | Bot | Visits | Last Seen | |-----|--------|-----------| | GPTBot | 12.4K | 2s ago | | PerplexityBot | 6.2K | 5s ago | | ClaudeBot | 2.1K | 12s ago | | Google-Extended | 4.1K | 3s ago | | CCBot | 1.8K | 8s ago | ### Page Coverage See which pages bots are indexing. Ensure important content is crawled. | Page | Coverage | Status | |------|----------|--------| | /pricing | 95% | indexed | | /features | 88% | indexed | | /docs | 72% | indexed | | /blog | 65% | indexed | | /api | 58% | indexed | ### Content Extraction Understand what content bots are extracting and how it's being used. #### Request Log - GET /api/products → 200 OK | 2.3KB - GET /api/pricing → 200 OK | 1.8KB - GET /docs/intro → 200 OK | 4.2KB #### Live Request Monitoring Track what content AI bots extract in real-time. ### Crawl Timeline View historical crawl data. Spot trends and anomalies over time. - GPTBot: 12.4K visits (trending up over 6 months) - PerplexityBot: 6.2K visits (trending up over 6 months) ### Click to Revenue Track how AI-driven clicks convert to actual revenue. Full attribution funnel. | Stage | Value | Percentage | |-------|-------|-----------| | AI Clicks | 12.4K | 100% | | Page Views | 8.2K | 66% | | Engaged | 3.1K | 25% | | Converted | 847 | 7% | **Attributed Revenue: $24,680** ### Prompt Injection Inject content and watch AI crawlers pick it up in real-time. Test your Actions. #### Live Detection Feed | Content | Bot | Status | Time | |---------|-----|--------|------| | /pricing-guide | GPTBot | detected | 2s ago | | /product-faq | PerplexityBot | crawling | 5s ago | | /api-docs | ClaudeBot | indexed | 12s ago | | /use-cases | Google-Extended | detected | 18s ago | --- ## Traffic Trend Charts ### AI Traffic vs Human Traffic (Monthly) | Month | AI Traffic | Human Traffic | |-------|-----------|---------------| | Jan | 2,400 | 65,000 | | Feb | 3,200 | 68,000 | | Mar | 4,100 | 71,000 | | Apr | 5,800 | 74,000 | | May | 7,200 | 78,000 | | Jun | 9,100 | 80,000 | | Jul | 11,400 | 82,000 | | Aug | 14,200 | 84,000 | | Sep | 16,800 | 86,000 | | Oct | 19,400 | 87,500 | | Nov | 22,100 | 88,500 | | Dec | 24,800 | 89,200 | ### Bot Distribution | Bot | Share | Visits | |-----|-------|--------| | ChatGPT | 49.8% | 12,400 | | Perplexity | 25.0% | 6,200 | | Gemini | 16.5% | 4,100 | | Claude | 5.7% | 1,400 | | Others | 3.0% | 700 | ### Session Quality Metrics | Metric | Value | Change | |--------|-------|--------| | AI Sessions | 24.8K | +156% MoM | | Avg Duration | 4:32 | +12% MoM | | Bounce Rate | 24% | -8% MoM | | Conv. Rate | 4.2% | +1.4% MoM | ### Conversion Attribution by Source | Source | Conversions | Rate | |--------|------------|------| | ChatGPT | 342 | 2.8% | | Perplexity | 186 | 3.0% | | Google AI | 145 | 1.9% | | Claude | 89 | 6.4% | | Direct | 85 | 1.1% | ### Source Platform Visits | Platform | Visits | Share | |----------|--------|-------| | ChatGPT | 12.4K | 49.8% | | Perplexity | 6.2K | 25.0% | | Gemini | 4.1K | 16.5% | | Claude | 2.1K | 8.5% | | Copilot | 1.8K | 7.2% | ### Traffic by Time - Peak: Wednesday 3PM EST --- ## Integrations — Works with your stack ### Cloudflare Edge detection at the network level. Intercept AI crawlers before they reach your origin. ### Vercel Native middleware integration. Track bot visits across serverless deployments. ### AWS Lambda@Edge functions for real-time bot classification with CloudWatch metrics. ### Fastly VCL snippets and Compute@Edge for cache-aware bot detection. ### Netlify Edge Functions integration for Jamstack sites and serverless endpoints. ### Akamai Enterprise-grade bot detection at the edge with Akamai EdgeWorkers integration. ### Google Cloud Cloud Functions and Cloud CDN integration for seamless bot analytics. ### WordPress One-click plugin install. Cache-compatible bot tracking with WP-CLI support. --- ## Testimonials > "We had no idea 18% of our traffic was coming from AI search until we installed Rankly. Now it's our fastest growing channel." > — James Wright, VP Marketing, SaaSCo > "The query-level insights are incredible. We can see exactly what questions lead people to us and optimize our content accordingly." > — Amanda Liu, Content Director, MediaHub > "Rankly paid for itself in the first week. We found conversion rates from AI traffic were 2x higher than organic search." > — David Okonkwo, Growth Lead, FinTechApp --- ## Start Tracking **Free Setup** ### Win every AI shopping conversation [Book a Demo](https://cal.com/rankly/30min) # Protocol Tracker > Live tracker for agentic commerce protocols. Every spec change, GitHub PR, and release — pulled directly from upstream repos, sorted by actual merge time, no LLM rewrites. Updated continuously. URL: https://www.tryrankly.com/protocol-tracker ## What this is A single feed of every merged pull request, release, and spec change across the protocols that govern AI commerce. Data comes verbatim from GitHub — titles, merge dates, authors, SHAs are exactly what's on the upstream repo, with light cleanup (conventional-commit prefixes stripped, `(#NNN)` suffixes removed) so the feed reads like product release notes instead of git log. ## Tracked protocols (16, across 4 categories) ### Commerce (3) - **ACP** — Agentic Commerce Protocol (OpenAI + Stripe). Repo: agentic-commerce-protocol/agentic-commerce-protocol. Launched September 29, 2025. - **UCP** — Universal Commerce Protocol (Google + Shopify, with Etsy, Wayfair, Target, Walmart and 20+ partners). Repo: Universal-Commerce-Protocol/ucp. Launched January 11, 2026. - **NVIDIA RAC** — Retail Agentic Commerce Blueprint. NVIDIA's open reference implementation of ACP + UCP. Repo: NVIDIA-AI-Blueprints/Retail-Agentic-Commerce. Launched January 5, 2026. ### Payments (7) - **AP2** — Agent Payments Protocol (Google). Repo: google-agentic-commerce/AP2. Launched September 16, 2025. - **x402** — Payments for HTTP (Coinbase, now Linux Foundation x402 Foundation). Repo: x402-foundation/x402. Launched May 6, 2025. - **MPP** — Machine Payments Protocol (Stripe + Tempo). Repo: tempoxyz/mpp-specs. Mainnet launch March 18, 2026. - **Visa TAP** — Trusted Agent Protocol (Visa). Repo: visa/trusted-agent-protocol. Launched October 2025. - **Visa ICC** — Visa Intelligent Commerce (broader Visa AI agent platform). Repo: visa/ai. Launched April 8, 2026. - **Verifiable Intent** — Mastercard + Google cryptographic delegation spec. Repo: agent-intent/verifiable-intent. Launched March 2026. - **L402** — Lightning HTTP 402 (Lightning Labs). Repo: lightninglabs/L402. Production since April 2020. ### Connectors (2) - **MCP** — Model Context Protocol (Anthropic + Linux Foundation AAIF). Repo: modelcontextprotocol/modelcontextprotocol. Launched November 25, 2024. - **A2A** — Agent2Agent Protocol (Google → Linux Foundation, 150+ orgs). Repo: a2aproject/A2A. Launched April 9, 2025. ### Discovery — AEO / GEO / AXP (4) - **NLWeb** — Natural Language Web (Microsoft). Repo: nlweb-ai/NLWeb. Launched May 19, 2025. - **WebMCP** — W3C draft bringing MCP to the browser. Repo: webmachinelearning/webmcp. Launched August 12, 2025. - **llms.txt** — Answer.AI's /llms.txt site index spec. Repo: AnswerDotAI/llms-txt. Launched September 2024. - **Web Bot Auth** — IETF draft for cryptographic bot identity (Cloudflare). Repo: cloudflare/web-bot-auth. Launched July 2025. ## How to use it - The protocol tracker page groups protocols into 4 tabs: Commerce, Payments, Connectors, Discovery. - Each card shows: provider favicons, the protocol's full name, a sparkline of merge activity over the last 14 days, total merged PR count, the latest merged PR (verbatim title + actual merge date + author), and a recent-PR list. Click any PR row to open a side panel with the full PR body, labels, commit SHA, and a "View on GitHub" link. - "Show all N merged PRs" opens a side panel with the protocol's most recent 50 merged PRs. - Provider favicons hover to reveal the company name. - The page also offers an email subscription for "alert me when a tracked protocol publishes a new PR." ## Why we built it Tracking 16 protocols across 4 categories by hand means watching ~20 GitHub repos, 8 spec websites, and 10+ vendor blogs. Engineers and PMs working on agentic commerce shouldn't need to do that — a single feed makes it possible to scan "what shipped this week across the agentic-commerce stack" in 30 seconds. ## Related - [UCP/ACP Validator](https://www.tryrankly.com/ucp-validator) — scan any merchant domain for UCP and ACP compliance. - [Agent Directory](https://www.tryrankly.com/agent-directory) — directory of 4,000+ AI agents, crawlers, and bots that visit websites. - [Rankly blog](https://www.tryrankly.com/blogs) — long-form analysis of the agentic commerce landscape. # Query Fanouts # Analysis History View your past query fanout analyses --- ## Header - Back arrow link to /try-query-fanouts-extension - Rankly logo and brand name - User email display --- ## Empty State ### No Analyses Yet Capture your first query fanout analysis using the browser extension. [Get Started](/try-query-fanouts-extension) --- ## Analysis List Each analysis entry displays: - **Date**: Formatted as month, day, year, hour, and minute - **Preview**: Chat title, query preview, or "No preview available" (links to /try-query-fanouts-extension/report?id={analysis_id}) - **Query Count**: Number of queries captured (with search icon) - **Citation Count**: Number of citations captured (with link icon) - **Delete Button**: Appears on hover, confirms before deleting. Disables the analysis via PATCH to /analyses/{id}/disable. ### Load More If more analyses are available, a "Load More" button appears to paginate results (20 per page). --- ## Error State If loading fails, displays the error message with a "Try Again" button. --- ## Footer Powered by Rankly | AEO Analytics Platform --- **URL:** https://www.tryrankly.com/try-query-fanouts-extension/history # Query Fanouts Report ## Header - Back arrow link to /try-query-fanouts-extension - Rankly logo and brand name - History button (links to /try-query-fanouts-extension/history, visible when logged in) - Export CSV button - Export JSON button --- ## Signup Modal — Unlock Full Analysis Displays query count and citation count. ### Unlock Full Analysis Enter your work email to continue — free forever. [Email input: you@company.com] [View Report] Skip for now (limited view) - On submit, creates/updates lead via POST to /leads - Saves analysis data via POST to /analyses (includes queries, citations, products, chatGptAnswer, userPrompt, chatTitle, contentType) - Stores email in sessionStorage and localStorage - Dispatches "rankly-store-email" event to extension --- ## Locked Section Component For users who skipped signup, sections display as locked: - Section title with count - "Locked" label with lock icon - Clicking opens signup modal --- ## No Data Captured State ### No Data Captured Make sure you've asked ChatGPT a question while the extension is capturing. [Go Back](/try-query-fanouts-extension) --- ## Title Section - Label: "Query Fanouts Report" - Title: Chat title from the analysis - Date: Current date formatted as month, day, year - Limited view indicator (if not logged in) - Conversation ID with copy button and external link to https://chatgpt.com/c/{conversationId} --- ## Stats Grid Displays key metrics in a bento-style grid: - **Queries**: Total number of query fanouts captured - **Citations**: Total number of source citations - **Products**: Total number of products listed (shown only if products > 0) - **Domains**: Number of unique domains cited --- ## Original User Query Displays the original question the user asked ChatGPT, with a copy button. --- ## Query Fanouts Section Expandable/collapsible section showing all captured query fanouts. Each query is numbered (01, 02, etc.) and has a copy button that appears on hover. --- ## Products Listed Section Expandable/collapsible section for product recommendations captured from ChatGPT's Shopping Connector. ### Product Carousel Products are displayed in a scrollable carousel with navigation arrows. Each product card shows: - Product image (square aspect ratio) - Product title (2-line clamp) - Price - Merchant name Clicking a product card selects it and highlights it with a blue border. ### Selected Product Detail Panel Three-column layout showing detailed information about the selected product: #### Overview Column - **AI Rerank**: Display position in the carousel (e.g., #1) with delta indicator showing position change (up/down arrow with number) - **Source Position**: Original citation position with cite reference (e.g., turn0product0) - **Rating**: Star rating and review count - **Listing Type**: MULTI-SELLER, SINGLE-IN-CLUSTER, or STANDALONE #### Search Context Column - **Market**: Country code / language code (e.g., IN / EN) - **Location**: Decoded from Google UULE location code to human-readable location - **Query**: The shopping search query used (URL-decoded, plus signs replaced with spaces) #### Data Sources Column - **Provider**: Google Shopping (provider code p2) - **GPCID**: Google Product Category ID - **Merchant ID**: Merchant identifier - **Catalog ID**: Product catalog identifier --- ## Sources Cited Section Expandable/collapsible section (locked for anonymous users). ### Filter Options - All (total count) - News (count of news citations) - Search (count of search citations) - With Date (count of citations with publication dates) ### Citation List Each citation displays: - Index number (01, 02, etc.) - Title (or domain if no title) - Reference type tag (news, search) - Publication date (if available, formatted as month day, year) - Full URL - Snippet (2-line clamp, if available) - External link icon on hover --- ## Source Types Section Expandable/collapsible section (locked for anonymous users). Grid of source type cards, each showing: - Count of citations - Source type label --- ## Top Cited Domains Section Expandable/collapsible section (locked for anonymous users). Ranked list (top 10) of most-cited domains, each showing: - Rank number (01, 02, etc.) - Domain name - Citation count - External link to domain --- ## ChatGPT Response Section Expandable/collapsible section showing the full ChatGPT response text with: - Character count - Copy button - Scrollable pre-formatted text (max height 384px) --- ## CTA Section — Scale Your Analysis ### Track Query Fanouts at Scale Monitor 100s of queries, track citation share over time, and get actionable AEO insights with Rankly. [Book a Demo](Calendly link) | [Learn More](https://tryrankly.com) --- ## Footer Powered by Rankly | AEO Analytics Platform --- ## Data Loading Behavior - If URL has `id` parameter: loads analysis from database via GET /analyses/{id} - If URL has `sessionId` and `queries` parameters: loads from sessionStorage (extension flow) with retry polling (up to 50 retries at 100ms intervals) - Listens for "extension-data-ready" custom event for real-time data from extension - If URL has `preview=1`: displays mock/demo data with sample products and queries - Auto-saves analysis for returning users (email in URL, no existing ID) via POST to /analyses ## Export Functionality - **CSV Export**: Exports queries and citations to CSV format - **JSON Export**: Exports queries, citations, chatGptAnswer, and conversationId to JSON format - Export requires user to be logged in (opens signup modal if not) --- **URL:** https://www.tryrankly.com/try-query-fanouts-extension/report # Query Fanouts — Signup This page is a redirect. It automatically redirects users to the report page at /try-query-fanouts-extension/report where the signup modal handles the registration flow. ## Redirect Behavior - Reads `sessionId` and `queries` parameters from the URL - Redirects to: `/try-query-fanouts-extension/report?sessionId={sessionId}&queries={queries}` - Displays a loading spinner during the redirect --- **URL:** https://www.tryrankly.com/try-query-fanouts-extension/signup # Query Fanouts # See What ChatGPT Searches Behind Your Queries Capture the internal search queries and citations AI uses to answer you. [Add to Chrome](https://chromewebstore.google.com/detail/aehjppkoldacichgkjomekhlohfdhkcm?utm_source=item-share-cb) [Open ChatGPT](https://chatgpt.com) Extension detected! Open ChatGPT to start capturing. --- ## Quick Start ## How It Works Get started in under 2 minutes with these simple steps. ### 01 — Install Extension Add the extension to Chrome from the Chrome Web Store. One click install. - Click "Add to Chrome" - Confirm the installation - Pin it for easy access [Add to Chrome](https://chromewebstore.google.com/detail/aehjppkoldacichgkjomekhlohfdhkcm?utm_source=item-share-cb) ### 02 — Ask ChatGPT Go to ChatGPT and ask any question. The extension captures queries and citations in real-time. "Best CRM tools for small businesses in 2025" Pro tip: Current events, products, or comparisons trigger the most searches. ### 03 — View Analysis Click "View Report" in the extension to see all captured data and insights. - Query fanouts (3-5 per question) - Source citations & domains - Export to CSV/JSON --- ## Features ## What You'll Capture ### 01 — Query Fanouts See every search query ChatGPT generates internally. ### 02 — Source Citations Track which websites ChatGPT cites - understand what content ranks in AI answers. ### 03 — Citation Analysis See citation types (Search, News) and domain authority patterns at a glance. ### 04 — Product Recommendations Capture products from ChatGPT's Shopping Connector with hidden metadata like cluster IDs and merchant info. --- ## Resources ## Learn More Simple, data-led explanations of AEO and how to improve it. [Enter your work email] [Get in touch] Thanks! We'll be in touch soon. ### Blog Card **Product** | 26 December #### Query Fanouts: See What Answer Engines Look For Discover how LLMs break down your queries into multiple search queries called query fanouts. [Read More](/blogs/query-fanouts) --- **URL:** https://www.tryrankly.com/try-query-fanouts-extension # Privacy Policy **Last updated: May 13, 2026** URL: https://www.tryrankly.com/privacy-policy --- ## 1. Introduction This Privacy Policy describes how Rankly Inc. ("Rankly," "we," "us," or "our") collects, uses, stores, and protects information when you access or use our services, including: - **Rankly Platform** (app.tryrankly.com) - Our Agentic Commerce Experience platform, a set of tools designed to optimize agentic, AI, and LLM traffic for brands and ecommerce websites - **Query Fanouts Chrome Extension** - A complimentary tool for analyzing AI-generated search behavior - **Marketing Website** (tryrankly.com) Rankly is committed to maintaining the highest standards of data privacy, transparency, and regulatory compliance. We operate under a **full data transparency policy**: no user data is processed, shared, or utilized for any purpose without explicit user consent. By using our services, you agree to the collection and use of information in accordance with this policy. If you do not agree with any part of this policy, please discontinue use of our services. ## 2. Information We Collect ### 2.1 Account Information When you create an account, we collect: - **Email Address** - Used for account identification, authentication, and service communications - **Name** - First and last name (optional, provided at your discretion) - **Company Name** - Your organization name (optional, provided at your discretion) - **Authentication Data** - If you sign in via Google OAuth, we receive your Google account identifier and basic profile information (name and email) ### 2.2 Agentic Commerce Experience Data Our analytics services collect and process the following: #### Brand Analysis Configuration: - **URLs** - Website URLs you submit for brand visibility analysis - **Brand Context** - Your brand name, industry classification, and value propositions - **Topics and Personas** - Search topics and user personas you configure for tracking - **Competitors** - Competitor brands you designate for comparative analysis #### AI Platform Monitoring Data: Rankly monitors your brand's visibility across mainstream AI assistants and agents. **All data collected from AI platforms is publicly available information**, gathered through the following process: - We query AI platforms using brand-relevant search prompts configured by you - All queries are executed in **anonymous, unauthenticated sessions** - no user accounts or personal credentials are used - We collect the AI-generated responses, including brand mentions, positioning, sentiment indicators, and source citations - This data is **identical to what any member of the public would receive** when submitting the same queries to these platforms **AI Platforms Currently Monitored:** OpenAI ChatGPT, Perplexity, and Google Search. We are actively integrating Google Gemini and Microsoft Copilot. We exclusively partner with AI platforms that comply with applicable United States federal and state regulations, European Union data protection laws (including GDPR), and the Republic of India's data protection legislation. We do not integrate with, transmit data to, or receive data from any AI platform that does not meet these compliance standards. ### 2.3 Google Analytics 4 (GA4) Integration When you connect your Google Analytics 4 account to Rankly, the following applies: - **Authorization** - We use Google OAuth 2.0 with PKCE (Proof Key for Code Exchange) to obtain your explicit consent before accessing any analytics data - **Scope of Access** - We request **read-only access** (`analytics.readonly`) to your GA4 property. We cannot and do not modify, delete, or write any data to your Google Analytics account - **Data Retrieved** - We read standard GA4 metrics including sessions, page views, user counts, engagement rates, traffic sources, geographic summaries, and device categories - **Purpose** - GA4 data is used exclusively to display your website traffic analytics within the Rankly dashboard, to identify traffic originating from AI assistants and agents, and to calculate derived metrics such as session quality scores and traffic source comparisons - **Data Processing** - We process GA4 data to classify traffic sources, detect AI-referral traffic patterns, compute period-over-period trends, and generate aggregated visual reports. All processed data is displayed solely to you, the authenticated account owner - **No Onward Transfer** - Your GA4 data is never shared with, sold to, or transmitted to any third party. It is not used for advertising, profiling, or any purpose other than displaying analytics within your Rankly dashboard - **Revocation** - You may disconnect your GA4 account at any time through the Rankly platform settings, or revoke Rankly's access directly via your Google Account at [https://myaccount.google.com/permissions](https://myaccount.google.com/permissions). Upon disconnection, all cached GA4 data is permanently deleted from our systems #### Google API Services User Data Policy / Limited Use Disclosure Rankly's use and transfer to any other app of information received from Google APIs will adhere to the [Google API Services User Data Policy](https://developers.google.com/terms/api-services-user-data-policy), including the **Limited Use** requirements. Specifically, with respect to data obtained via the `https://www.googleapis.com/auth/analytics.readonly` scope and the `userinfo.email` / `userinfo.profile` scopes: - We use this data **solely** to provide and improve user-facing features of the Rankly platform - namely, to display the user's own Google Analytics 4 traffic data (broken down by AI platform referrers such as ChatGPT, Perplexity, Gemini, and Google AI Overviews) inside the user's authenticated Rankly dashboard - We **do not** transfer this data to others, except as necessary to provide or improve user-facing features, comply with applicable law, or as part of a merger, acquisition, or sale of assets with prior notice to affected users - We **do not** use or transfer this data for serving advertisements, including retargeting, personalized, or interest-based advertising - We **do not** allow humans to read this data, except: (a) with the user's affirmative consent for specific individual messages; (b) where necessary for security purposes, such as investigating a bug or abuse; (c) to comply with applicable law; or (d) where the data has been aggregated and anonymized and is used for internal operations - Tokens issued under Google OAuth are stored encrypted at rest, used only on the user's behalf to query their own GA4 properties, and revoked immediately upon user disconnection ### 2.4 Query Fanouts Chrome Extension The Query Fanouts extension captures the following data only when you actively initiate a capture session: - **Search Queries** - The internal search queries generated by ChatGPT during web search operations - **Citations** - URLs and metadata of sources referenced by ChatGPT - **Product Data** - Product information displayed in shopping recommendation panels - **Search Context** - Geographic location context (UULE codes) and language settings **We do not capture your personal prompts, private conversations, or any content you type into ChatGPT.** We exclusively capture the search queries that ChatGPT generates internally as part of its web retrieval process. ### 2.5 Usage and Billing Data - **Service Usage Metrics** - API token counts, response times, and feature utilization statistics used for service optimization - **Subscription Information** - Your plan tier and billing cycle - **Payment Information** - Payment processing is handled exclusively by our PCI-compliant payment processor. We do not store, process, or have access to your credit card numbers, bank account details, or other financial instrument data ## 3. How We Use Your Information We use collected information strictly for the following purposes: - To operate, maintain, and deliver our analytics services - To analyze and display your brand's visibility across AI assistants and agents - To process and display your GA4 traffic analytics in read-only dashboards - To generate insights, trend analyses, and actionable reports - To provide data export functionality (CSV/JSON) - To process subscription payments and manage billing - To communicate with you regarding your account, service updates, and support requests - To improve service reliability, performance, and user experience **We do not use your data for advertising, behavioral profiling, or sale to any third party under any circumstances.** ## 4. Data Sovereignty and International Compliance ### 4.1 Infrastructure and Data Residency All Rankly infrastructure is hosted on **Google Cloud Platform (GCP)** within the **United States (us-east1 region)**. Our infrastructure components include: - **Application Services** - Google Cloud Run (serverless compute) - **Database Systems** - Self-hosted on Google Compute Engine within GCP's US data centers - **Data Backups** - Google Cloud Storage within the US region ### 4.2 Regulatory Compliance Rankly is designed and operated to comply with: - **United States** - Applicable federal and state data protection regulations, including the California Consumer Privacy Act (CCPA) and the California Privacy Rights Act (CPRA) - **European Union** - General Data Protection Regulation (GDPR), including provisions for lawful processing, data minimization, and data subject rights - **Republic of India** - The Digital Personal Data Protection Act, 2023 (DPDPA), including provisions for data principal rights and cross-border data transfer requirements ### 4.3 Prohibited Data Transfers Rankly maintains a strict policy against transmitting, mirroring, replicating, or otherwise transferring any user data - whether personal, analytical, or behavioral - to any jurisdiction, service provider, or platform that does not comply with the regulatory frameworks enumerated in Section 4.2. We do not engage with service providers domiciled in, or routing data through, jurisdictions that lack adequate data protection standards as assessed under GDPR adequacy decisions, US regulatory guidance, or equivalent frameworks. ### 4.4 Third-Party AI Platform Compliance All AI platforms integrated with Rankly are evaluated against the following criteria before integration: - Compliance with US federal and state data protection regulations - Compliance with GDPR and EU data protection standards - Compliance with Indian data protection legislation - Transparent data handling and privacy policies - No evidence of unauthorized data sharing or onward transfer to non-compliant jurisdictions Platforms that fail to meet any of these criteria are not integrated, regardless of market demand or technical capability. ## 5. Third-Party Service Providers We engage the following categories of third-party service providers, each bound by contractual data protection obligations: | Provider Category | Purpose | Data Shared | |---|---|---| | **Cloud Infrastructure** (Google Cloud Platform) | Application hosting, compute, storage | All service data (encrypted at rest and in transit) | | **AI Platform Providers** (OpenAI, Google, Perplexity) | Brand visibility monitoring via public queries | Anonymized search queries only; no user personal data | | **Payment Processor** (Dodo Payments) | Subscription billing and payment processing | Email address and subscription tier; no analytics data | | **Email Service** (SMTP via Google Workspace) | Transactional and account communications | Email address and message content | | **Google OAuth / GA4 API** | User authentication and analytics data retrieval | OAuth tokens (encrypted); GA4 data displayed only to the authenticated user | **No user data - including personal information, GA4 analytics, brand analysis results, or account details - is shared with any entity not listed above.** Each provider processes data solely for the stated purpose and is prohibited from using it for independent purposes. ## 6. Data Sharing Policy **We do not sell, rent, lease, trade, or otherwise transfer your personal information or analytics data to any third party for commercial, advertising, or any other purpose.** Data may be disclosed only in the following limited circumstances: - **Service Operations** - To the third-party service providers listed in Section 5, strictly for the purpose of delivering our services, under binding data protection agreements - **AI Platform Queries** - Search prompts submitted to AI platforms for brand monitoring are anonymized and contain no personally identifiable information - **Legal Obligations** - If required by law, regulation, subpoena, court order, or governmental request from a jurisdiction with lawful authority - **Protection of Rights** - To protect the rights, property, or safety of Rankly, our users, or the public, as permitted by law - **Aggregated Statistics** - In anonymized, aggregated form that cannot be used to identify any individual user, for industry research or statistical reporting purposes ## 7. Data Security We implement comprehensive technical and organizational security measures, including: - **Encryption in Transit** - All data transmitted between your browser and our servers is encrypted using TLS 1.2 or higher - **Encryption at Rest** - All stored data is encrypted using industry-standard encryption algorithms - **Access Controls** - Role-based access controls and principle of least privilege for all system access - **Authentication Security** - OAuth 2.0 with PKCE for third-party integrations; JWT-based session management with secure token handling - **Infrastructure Security** - Firewall rules, network segmentation, and regular security assessments - **Credential Security** - API keys and secrets are stored in environment variables and never exposed in client-side code While we employ commercially reasonable measures to protect your data, no method of electronic transmission or storage is entirely infallible. We cannot guarantee absolute security but are committed to promptly addressing any security incidents in accordance with applicable breach notification laws. ## 8. Data Retention - **Account Data** - Retained for the duration of your active account and for thirty (30) days following account deletion to facilitate recovery requests, after which it is permanently deleted - **Analytics and Brand Monitoring Data** - Retained while your account is active to provide historical tracking and trend analysis. Deleted within thirty (30) days of account termination - **GA4 Cached Data** - Cached for up to five (5) minutes for performance optimization. All cached GA4 data is permanently deleted upon disconnection of your GA4 account - **Chrome Extension Data (Local)** - Stored exclusively in your browser's local storage until you clear it. We do not retain extension data on our servers unless you explicitly save a report - **Payment Records** - Retained as required by applicable tax and financial regulations ## 9. Your Rights Regardless of your jurisdiction, we honor the following data subject rights: - **Right of Access** - Request a copy of all personal data we hold about you - **Right to Deletion** - Request permanent deletion of your account and all associated data - **Right to Rectification** - Request correction of any inaccurate personal data - **Right to Data Portability** - Export your data at any time using our built-in CSV and JSON export functionality - **Right to Withdraw Consent** - Revoke consent for any data processing, including disconnecting third-party integrations (e.g., GA4), at any time without affecting the lawfulness of processing performed prior to withdrawal - **Right to Object** - Object to specific forms of data processing - **Right to Restrict Processing** - Request restriction of processing in certain circumstances To exercise any of these rights, contact us at **privacy@tryrankly.com** or **contact@tryrankly.com**. We will respond to all requests within thirty (30) days, or within the timeframe required by applicable law. ## 10. Cookies and Tracking Technologies - **Essential Cookies** - Used for authentication, session management, and security (e.g., GA4 OAuth session cookies) - **Preference Cookies** - Used to remember your dashboard settings and display preferences - **Analytics** - We may use first-party analytics to understand aggregate usage patterns. We do not use third-party advertising trackers or behavioral profiling cookies The Query Fanouts Chrome Extension does not use cookies and only activates on ChatGPT pages when you explicitly initiate a capture session. ## 11. Children's Privacy Our services are not directed to individuals under the age of sixteen (16). We do not knowingly collect personal information from children under 16. If we become aware that we have inadvertently collected such data, we will promptly delete it. ## 12. Changes to This Policy We may update this Privacy Policy periodically to reflect changes in our practices, services, or applicable law. Material changes will be communicated by posting the revised policy on this page, updating the "Last updated" date, and where appropriate, notifying you via email. Your continued use of our services following the posting of changes constitutes your acceptance of the revised policy. ## 13. Contact Us For questions, concerns, or requests related to this Privacy Policy or our data practices, please contact us at: - **Privacy Inquiries:** privacy@tryrankly.com - **General Support:** contact@tryrankly.com - **Website:** tryrankly.com # Terms of Service **Last updated: March 18, 2026** URL: https://www.tryrankly.com/terms-of-service --- ## 1. Acceptance of Terms By accessing or using any services provided by Rankly Inc. ("Rankly," "we," "us," or "our"), including the Rankly platform at app.tryrankly.com, the Query Fanouts Chrome Extension, and the marketing website at tryrankly.com, you acknowledge that you have read, understood, and agree to be bound by these Terms of Service ("Terms") and our Privacy Policy. If you do not agree to these Terms in their entirety, you must not access or use our services. If you are using our services on behalf of a legal entity, you represent and warrant that you have the authority to bind that entity to these Terms. ## 2. Description of Services ### 2.1 Rankly Platform The Rankly platform (app.tryrankly.com) is an Agentic Commerce Experience platform: a set of tools designed to optimize agentic, AI, and LLM traffic for brands and ecommerce websites. It enables businesses to: - Monitor and analyze brand visibility across AI assistants and agents, including OpenAI ChatGPT, Perplexity, and Google Search, with active integration of Google Gemini and Microsoft Copilot - Track citation sources, sentiment indicators, and competitive positioning in AI-generated responses - Measure AI agent and crawler traffic on their own websites, alongside human traffic - Connect Google Analytics 4 (GA4) accounts for read-only traffic analytics, including AI-referral traffic detection and attribution - Generate actionable insights through data aggregation, trend analysis, and cross-platform performance reporting **AI Platform Compliance:** Rankly exclusively integrates with AI platforms that comply with United States federal and state regulations, European Union data protection laws (including GDPR), and the Republic of India's data protection legislation (including DPDPA 2023). We do not integrate with any AI platform that fails to meet these regulatory standards. **Public Data Collection:** All brand monitoring data collected from AI assistants and agents is publicly available information. Queries are executed in anonymous, unauthenticated sessions. No user credentials, personal data, or proprietary information is transmitted to any AI platform during the monitoring process. ### 2.2 Google Analytics 4 Integration Rankly offers optional GA4 integration subject to the following terms: - **Read-Only Access** - Rankly requests and operates under the `analytics.readonly` OAuth scope. We cannot and do not modify, delete, or write any data to your Google Analytics account - **User-Initiated Connection** - GA4 integration requires your explicit authorization via Google OAuth 2.0. You may revoke this authorization at any time - **Data Usage** - GA4 data is processed solely to display traffic analytics, detect AI-referral traffic patterns, and generate performance reports within your Rankly dashboard - **No Onward Transfer** - GA4 data is never shared with, sold to, or transmitted to any third party. It is displayed exclusively to you, the authenticated account owner - **Data Deletion** - Upon disconnecting your GA4 account, all cached and stored GA4 data is permanently deleted from our systems ### 2.3 Query Fanouts Chrome Extension The Query Fanouts Chrome Extension is a complimentary tool that enables users to capture and analyze the internal search queries, citations, and product recommendations generated by ChatGPT during web search operations. This extension is provided at no cost as a companion tool to the Rankly platform. The extension captures only ChatGPT's internally generated search queries. It does not capture, record, or transmit your personal prompts, conversation content, or any data you input into ChatGPT. ## 3. Account Registration and Security To access certain features, you must create an account. By registering, you agree to: - Provide accurate, current, and complete registration information - Maintain and promptly update your information to keep it accurate - Maintain the confidentiality and security of your account credentials - Notify us immediately at contact@tryrankly.com of any unauthorized access to or use of your account - Accept sole responsibility for all activities that occur under your account Rankly reserves the right to suspend or terminate accounts that contain inaccurate information or that are used in violation of these Terms. ## 4. Data Transparency and User Rights Rankly operates under a **full data transparency policy**: - **No Hidden Processing** - All data processing activities are disclosed in our Privacy Policy. No user data is processed for any purpose not explicitly stated therein - **No Data Sale** - We do not sell, rent, lease, or trade user data to any third party under any circumstances - **No Advertising Use** - User data is never used for targeted advertising, behavioral profiling, or marketing by third parties - **No Unauthorized Sharing** - User data is shared only with the service providers identified in our Privacy Policy, solely for the purpose of delivering our services - **Full User Control** - You may export, modify, or delete your data at any time. You may disconnect third-party integrations (including GA4) at any time, triggering immediate deletion of associated cached data - **Jurisdictional Compliance** - All data processing complies with applicable US, EU (GDPR), and Indian (DPDPA 2023) data protection regulations. We do not transmit user data to any jurisdiction or entity that does not meet these compliance standards ## 5. Acceptable Use Policy You agree not to use our services to: - Violate any applicable law, regulation, or third-party right - Interfere with or disrupt the integrity, performance, or availability of our services or infrastructure - Attempt to gain unauthorized access to our systems, networks, or other users' accounts - Transmit malware, viruses, worms, or other harmful or malicious code - Scrape, harvest, or programmatically collect data from our services without prior written authorization - Use our services for any fraudulent, deceptive, or unlawful purpose - Reverse engineer, decompile, disassemble, or otherwise attempt to derive the source code of our services - Resell, sublicense, or redistribute our services or any data obtained through our services without authorization - Use our services in any manner that could damage, disable, or impair our infrastructure or interfere with other users' access Violation of this Acceptable Use Policy may result in immediate suspension or termination of your account without prior notice. ## 6. Intellectual Property ### 6.1 Rankly Property All content, features, functionality, software, designs, trademarks, logos, and other intellectual property comprising our services are owned by Rankly Inc. and are protected by applicable copyright, trademark, patent, and other intellectual property laws. No right, title, or interest in our intellectual property is transferred to you by virtue of your use of our services. ### 6.2 User Data You retain all ownership rights to the data you submit to our services. By using our services, you grant Rankly a limited, non-exclusive, non-transferable license to use, store, and process your data solely for the purpose of providing and improving our services. This license terminates upon deletion of your account or the relevant data. ### 6.3 Third-Party Content Data captured from third-party AI platforms (such as ChatGPT search queries via the Chrome extension or AI-generated responses during brand monitoring) is the intellectual property of the respective platform operators. Rankly does not claim ownership of such third-party content. You are responsible for ensuring your use of such data complies with the terms of the respective third-party services. ## 7. Subscription and Payment Terms ### 7.1 Rankly Platform Subscriptions - Certain platform features require a paid subscription. Plan details, pricing, and billing cycles are presented during the subscription process - Subscriptions renew automatically at the end of each billing period unless cancelled prior to the renewal date - You may cancel your subscription at any time. Cancellation takes effect at the end of the current billing period - Prices are subject to change with no less than thirty (30) days' prior notice - You are responsible for all applicable taxes, duties, and levies associated with your subscription ### 7.2 Payment Processing All payment transactions are processed by our PCI-compliant third-party payment processor. Rankly does not store, process, or have access to your credit card numbers, bank account details, or other financial instrument data. Payment disputes should be directed to contact@tryrankly.com. ### 7.3 Query Fanouts Extension The Chrome extension is provided free of charge. We reserve the right to introduce premium features in the future with appropriate notice. ## 8. Service Availability and Modifications - We strive to maintain continuous service availability but do not guarantee uninterrupted or error-free operation - We reserve the right to modify, update, or discontinue any feature or service with reasonable notice - Scheduled maintenance windows will be communicated in advance where practicable - We are not liable for any loss or damage resulting from service interruptions, whether planned or unplanned ## 9. Disclaimers OUR SERVICES ARE PROVIDED ON AN "AS IS" AND "AS AVAILABLE" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, NON-INFRINGEMENT, AND ACCURACY. We do not warrant that: (a) our services will meet your specific requirements; (b) our services will be uninterrupted, timely, secure, or error-free; (c) the results obtained from the use of our services will be accurate, complete, or reliable; or (d) any errors in our services will be corrected. AI-generated content monitored by our platform is produced by third-party AI systems. Rankly does not control, endorse, or guarantee the accuracy of such content. Analytics and insights provided by Rankly are informational in nature and should not be construed as professional, legal, or financial advice. ## 10. Limitation of Liability TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, RANKLY INC., ITS OFFICERS, DIRECTORS, EMPLOYEES, AGENTS, AND AFFILIATES SHALL NOT BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, EXEMPLARY, OR PUNITIVE DAMAGES, INCLUDING BUT NOT LIMITED TO LOSS OF PROFITS, REVENUE, DATA, GOODWILL, OR BUSINESS OPPORTUNITY, ARISING OUT OF OR RELATED TO YOUR USE OF OR INABILITY TO USE OUR SERVICES, REGARDLESS OF THE THEORY OF LIABILITY (CONTRACT, TORT, STRICT LIABILITY, OR OTHERWISE) AND EVEN IF RANKLY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Our total aggregate liability for all claims arising out of or relating to these Terms or your use of our services shall not exceed the greater of: (a) the total amounts you paid to Rankly in the twelve (12) months immediately preceding the event giving rise to the claim; or (b) one hundred United States dollars (USD $100). ## 11. Indemnification You agree to indemnify, defend, and hold harmless Rankly Inc. and its officers, directors, employees, and agents from and against any and all claims, liabilities, damages, losses, costs, and expenses (including reasonable attorneys' fees) arising out of or related to: (a) your use of our services; (b) your violation of these Terms; (c) your violation of any third-party rights; or (d) any content or data you submit through our services. ## 12. Termination ### 12.1 By You You may terminate your account at any time by contacting us at contact@tryrankly.com or through your account settings. Upon termination, your data will be retained for thirty (30) days to facilitate recovery requests, after which it will be permanently deleted. ### 12.2 By Rankly We may suspend or terminate your access to our services immediately, with or without notice, if we determine in our sole discretion that you have violated these Terms, engaged in fraudulent or abusive conduct, or if continued access poses a risk to our services, other users, or third parties. ### 12.3 Effect of Termination Upon termination: (a) your right to access and use our services ceases immediately; (b) any outstanding payment obligations remain due; and (c) Sections 6, 9, 10, 11, and 13 survive termination. ## 13. Governing Law and Dispute Resolution These Terms shall be governed by and construed in accordance with the laws of the State of Delaware, United States, without regard to its conflict of law provisions. Any dispute, controversy, or claim arising out of or relating to these Terms or your use of our services shall first be subject to good-faith negotiation between the parties for a period of thirty (30) days. If the dispute is not resolved through negotiation, it shall be resolved exclusively in the state or federal courts located in the State of Delaware, and you consent to the personal jurisdiction of such courts. ## 14. General Provisions - **Entire Agreement** - These Terms, together with the Privacy Policy, constitute the entire agreement between you and Rankly regarding your use of our services - **Severability** - If any provision of these Terms is found to be unenforceable, the remaining provisions shall continue in full force and effect - **Waiver** - Our failure to enforce any provision of these Terms shall not constitute a waiver of that provision or any other provision - **Assignment** - You may not assign or transfer your rights under these Terms without our prior written consent. Rankly may assign its rights and obligations without restriction - **Force Majeure** - Rankly shall not be liable for any failure or delay in performance due to circumstances beyond our reasonable control, including natural disasters, acts of government, or infrastructure failures ## 15. Contact Us For questions or concerns regarding these Terms of Service, please contact us at: - **Legal Inquiries:** legal@tryrankly.com - **General Support:** contact@tryrankly.com - **Website:** tryrankly.com