AI ecommerce statistics 2026: traffic, conversion, and agentic commerce

AI-referred visits to US retail sites converted 90% better than non-AI traffic in September 2026 in Adobe Analytics' panel, up from 49% worse in January 2025. Year-over-year traffic growth has fallen as the base has grown, to +71% in September 2026. AI referrals remain a small share of visits: 0.2% of Q4 2025 sessions in Contentsquare's global benchmark. Every figure below carries its source, sample and period.

Consumer survey terms

Adobe's year-over-year growth rate fell from +4,700% in July 2025 to +1,200% in October 2025, +269% in March 2026 and +71% in September 2026, as the prior-year base grew. Adobe does not publish AI-referred traffic as a share of total retail visits. The figures below come from other datasets and are not combined with Adobe's.

Share of traffic and click-through, other datasets

Each share has a different denominator: all web traffic across industries (Contentsquare), all sessions on a 1,898-site panel with organic ChatGPT only (Kaiser and Schulze), and retail ecommerce visits with no stated period (Similarweb). They are shown side by side, not as a trend.

Why the datasets differ

1. 01 · Different comparators

   Adobe measures AI referrals against a blended baseline that contains paid social. Kaiser and Schulze compare ChatGPT referrals with each channel individually. Shopify compares with organic search only. Beating a blend that includes paid social is not the same test as beating organic search.

2. 02 · Different periods

   The 973-site study covers February to August 2025. Adobe's own series was negative over the same months: 38% worse in April and 23% worse in July 2025. Adobe's readings turned positive from September 2025, and the 1,898-site follow-up finds ChatGPT conversion approximately matched organic search before February 2026.

3. 03 · Different populations

   Adobe is US retail only, weighted toward large enterprise retailers. The 973 sites are global and span 24 categories including finance, travel and jobs. Shopify covers its own merchants with no stated geography.

4. 04 · Different definitions of AI traffic

   Adobe counts referrals from chat services and AI browsers. Kaiser and Schulze isolate organic ChatGPT referrals. Shopify's metric wording changes each quarter.

5. 05 · Selection effects

   Peak-season AI referrals arrive with gift intent and a deadline. The 973-site study finds ChatGPT outcomes strongest in complex product categories. Each dataset may be sampling a different stage of the purchase journey.

No source has measured both populations on the same definition over the same window. Every 2026 reading of unpaid AI referrals on this page, from Adobe, Shopify and the 1,898-site panel, puts conversion at or above organic search or the blended baseline. Paid ChatGPT ad visits converted at 0.37% against 1.17% for paid Google Search. This page does not average the datasets.

Adobe Consumer Survey, n = 5,000+ US respondents, fielded July 2026

Other surveys, late 2025 to September 2026

Question wording, base and recall window differ between surveys, so the rows are not a ranking or a trend. The Omnisend figure is a vendor survey with unpublished question wording. OpenAI never published how many merchants offered Instant Checkout before dropping it in March 2026, and no other survey here reports a comparable share. The Menlo Ventures base is US shoppers who use AI, a subgroup whose size is not published.

Earlier Adobe surveys: n = 5,000 (published August 2025) and n = 1,000+ (Holiday 2025, published January 2026)

Adobe's three surveys have different sample sizes, fielding dates and question wording. The 85% (2025, n = 5,000), 81% (Holiday 2025, n = 1,000+) and 83% (July 2026, n = 5,000+) improved-experience figures are not a trend line.

Series sources: Salesforce, via Business Wire, 28 November 2023 · Salesforce, 4 December 2024 · Salesforce, 5 December 2025

Salesforce forecast $73 billion and 22% for Cyber Week 2025 on 20 November 2025 and reported $67 billion and 20% after it. The forecast window was 27 November to 1 December; the result window was 25 November to 1 December. The forecast is cited here only so the two are not confused. Salesforce, 20 November 2025

Retailer-owned assistants are a separate channel from third-party AI referrals: a shopper using Rufus or Sparky is already on the retailer's own site or app, so this usage is not an AI referral as Adobe and Shopify define one. Walmart's order value comparisons set Sparky users against non-users, who self-select, so they do not measure an effect of the assistant. Amazon's Alexa for Shopping combines Rufus and Alexa+, so its customer count does not continue the 2025 Rufus series. Amazon Investor Relations, 29 October 2026 · Amazon, 5 February 2026 · Walmart, 19 February 2026 · Walmart, 20 August 2026

Shopify describes its AI metric differently in each quarter's materials and does not say whether the definition changed, so the quarterly multiples are shown one quarter at a time. Shopify is one of the platforms in the ecommerce platform comparison. Shopify Investor Relations, 3 November 2026

Thanksgiving to Cyber Monday 2025

Prime Day 2025 and 2026

The peak-day conversion gaps for Thanksgiving and Black Friday 2025 are wider than the season average of +31%, which is consistent with the selection effect described in the conversion section: shoppers arriving from an AI service on those days carry gift intent and a deadline. Prime Day 2026 ran from 23 to 26 June and Prime Day 2025 from 8 to 11 July, so the year-over-year comparison spans different weeks. Adobe reports the 2025 season's total US online spend at $257.8 billion but does not publish AI-referred traffic as a share of it, and no share figure is estimated here. Adobe had not published measured figures for the 2026 holiday season at 9 November 2026; its forecast is in the forecasts table.

Sources: ACP from the ACP specification and OpenAI's merchant page (both retrieved 9 November 2026), OpenAI (24 March 2026) and Stripe (29 September 2025). UCP from Google (11 January, 19 March, 15 October and 4 November 2026), the UCP project (24 April 2026) and Merchant Center Help (retrieved 9 November 2026). AP2 from the AP2 specification (version 0.2) and Google (28 April 2026). A2A from the Agentic AI Foundation (17 August 2026). MCP from the MCP project (28 July 2026). MCP is included to show what it does not cover. OpenAI and Stripe, ACP repository, Retrieved 9 November 2026 (latest stable version 2026-04-17) · OpenAI, Retrieved 9 November 2026 · Universal Commerce Protocol project, 24 April 2026 · AP2 project, Retrieved 9 November 2026 (version 0.2, 28 April 2026) · Google Cloud, 16 September 2025 · Google, 28 April 2026 · Agentic AI Foundation, 17 August 2026 · Linux Foundation, 9 April 2026 · Model Context Protocol project, 28 July 2026 · Linux Foundation, 9 December 2025

What happened to in-chat checkout

Stripe and OpenAI launched Instant Checkout in ChatGPT on 29 September 2025, US only, with US-based Etsy businesses at launch and Shopify merchants described as coming soon. The launch introduced the Shared Payment Token, scoped to a specific merchant and cart total. Stripe, 29 September 2025 On 24 March 2026 OpenAI wrote: "We've found that the initial version of Instant Checkout did not offer the level of flexibility that we aspire to provide, so we're allowing merchants to use their own checkout experiences while we focus our efforts on product discovery." OpenAI, 24 March 2026

The best available evidence on why comes from Walmart. Its EVP Daniel Danker told Wired, as reported by Modern Retail, that conversion rates were three times lower for selection sold directly inside the chatbot than for products requiring a click out to the merchant. Etsy told the same outlet it did not see large volume from Instant Checkout but does see valuable referral traffic. This is a named-executive disclosure from one retailer, graded C, and it is the only conversion comparison of in-chat checkout against redirect that has been made public. Modern Retail, 27 March 2026

OpenAI has not published how many merchants went live on Instant Checkout, and counts reported by other outlets could not be traced to a primary source. OpenAI's merchant page now says: "We're moving away from a standalone Instant Checkout experience in ChatGPT and prioritizing better shopping discovery and merchant-owned checkout experiences." It also says "There are no fees on purchases that start in ChatGPT." OpenAI's developer documentation still describes Instant Checkout as available to approved partners, which conflicts with the merchant page. OpenAI, Retrieved 9 November 2026

What ACP is now

OpenAI describes ACP as "the connective layer between merchants and users throughout discovery," through which merchants share product feeds and promotions so their catalogs are represented in ChatGPT, with delivery paths that include third-party providers such as Salesforce and Stripe. OpenAI names Target, Sephora, Nordstrom, Lowe's, Best Buy, The Home Depot and Wayfair as integrated for discovery. Shopify merchants' product data is integrated through Shopify Catalog with no additional merchant work, and Shopify's Mani Fazeli is quoted saying buyers "can complete purchases on the merchant's online store through an in-app browser." Walmart has an in-ChatGPT app with account linking, loyalty and Walmart payments. The rollout covers free, Go, Plus and Pro users. OpenAI, 24 March 2026 The protocol itself still specifies checkout and delegated payment. Its latest stable version, 2026-04-17, covers cart, feed, orders, authentication and MCP, and the specification remains in beta, maintained by OpenAI and Stripe. OpenAI and Stripe, ACP repository, Retrieved 9 November 2026 (latest stable version 2026-04-17) The agentic commerce explainer covers the model in more depth.

What UCP is, and what is actually available

Google announced UCP at NRF on 11 January 2026 as an open-source standard co-developed with Shopify, Etsy, Wayfair, Target and Walmart, endorsed by more than 20 partners including Adyen, American Express, Best Buy, Flipkart, Macy's Inc, Mastercard, Stripe, The Home Depot, Visa and Zalando, and compatible with the Agent Payments Protocol, Agent2Agent and the Model Context Protocol as transports. Google Developers Blog, 11 January 2026 On 19 March 2026 Google added optional Cart and Catalog capabilities, the latter giving agents real-time pricing and inventory from retailer catalogs, and highlighted Identity Linking for loyalty benefits. Google, 19 March 2026 On 24 April 2026 the UCP Tech Council grew to 16 seats, adding Amazon, Meta, Microsoft, Salesforce and Stripe to members including Google, Shopify, Etsy, Target and Wayfair. Universal Commerce Protocol project, 24 April 2026

The specification and the checkout feature are different things. The specification is published and open. Google Merchant Center Help states that "the checkout feature enabled by UCP is available for select merchants at this time," that eligibility is limited to products in the United States, Canada and Australia, and that merchants express interest through a form. Retailers remain merchant of record. That wording was unchanged between 8 September and 9 November 2026. Google Merchant Center Help, Retrieved 9 November 2026

Google's consumer launches have moved ahead of that page. Universal Cart, with checkout powered by UCP, rolled out in the US across Google Search and the Gemini app on 15 October 2026, with Nike, Sephora, Target, Ulta Beauty, Walmart, Wayfair and many Shopify merchants named. Shoppers can check out with Google Pay or send the cart to the merchant's site. Google, 15 October 2026 On 4 November 2026 direct checkout powered by UCP began rolling out to Canadian shoppers in AI Mode and the Gemini app, and Google said simplified UCP integration will be available directly in Merchant Center from 2027. Google Canada, 4 November 2026 No merchant count or transaction volume has been published for either.

Where MCP fits

On 9 December 2025 Anthropic donated the Model Context Protocol to the Agentic AI Foundation, a directed fund under the Linux Foundation co-founded by Anthropic, Block and OpenAI with support from Google, Microsoft, AWS, Cloudflare and Bloomberg. MCP joined goose and AGENTS.md as founding projects, and its maintainers retain autonomy over technical direction. At donation the project reported 97 million+ monthly SDK downloads and 10,000 active servers. Model Context Protocol project, 9 December 2025 Anthropic, 9 December 2025 On 28 July 2026 it reported close to half a billion monthly downloads across its Tier 1 SDKs, a scope that may differ from the December figure. No newer server count has been published. Model Context Protocol project, 28 July 2026 MCP is not a commerce protocol. It defines how a model reaches tools and data, which is why UCP lists it as one transport among several, and it does not define carts, checkout or payment credentials. The MCP server directory lists the ecommerce servers that exist on it.

Agent-executed shopping: what has been measured

These are the first published measures of agents shopping on their own. Quantum Metric's figure comes from vendor platform data with no stated site count or period. Visa's is a laboratory benchmark that has not yet tested checkout or payment. Visa's European programme names issuers and merchants but no transaction count.

The two largest sponsors have taken different paths. OpenAI moved ChatGPT purchases to merchant-owned checkout in March 2026, and Walmart has reported in-chat conversion three times lower than click-out. Google launched checkout on its own surfaces in the US and Canada in October and November 2026, with Google Pay or a hand-off to the merchant's site. No conversion comparison for Google's checkout has been published.

1. 01 · Tag-based analytics reports human sessions by design

   Google Analytics 4 excludes traffic from known bots and spiders automatically, using Google research and the Interactive Advertising Bureau's International Spiders and Bots List, and the exclusion cannot be disabled or its volume seen. Any figure from a tag-based tool is therefore a human-session figure by construction. The requests that AI systems make to fetch product pages on a shopper's behalf, or to crawl a catalog, live only in server logs. Google Analytics Help, Retrieved 8 September 2026

2. 02 · User-agent strings are claims, not identity

   Google publishes a verification procedure because, in its words, site owners may be "concerned that spammers or other troublemakers are accessing your site while claiming to be from Google." The procedure is a reverse DNS lookup on the requesting IP, or a match against Google's published IP range files. Google Search Central, Retrieved 8 September 2026 OpenAI publishes separate IP address lists for OAI-SearchBot, GPTBot and ChatGPT-User for the same reason. OpenAI, Retrieved 8 September 2026 Allowlists and blocklists built on user-agent strings alone work from unverified identity data.

3. 03 · Attribution rules credit the last recorded touch

   Google Analytics 4 attribution models assign credit to touchpoints by rule or by a data-driven algorithm, and every model excludes direct visits from receiving credit unless the whole path is direct. Google Analytics Help, Retrieved 8 September 2026 A shopper who researches in an LLM without clicking through, then buys via a branded search or by typing the URL, leaves no AI touchpoint to credit. The purchase is attributed to search, or to whatever preceded the direct visit.

4. 04 · Some platforms expose no server logs at all

   The only unambiguous record of automated requests is the server or edge log, and not every hosted platform gives merchants one. Our server log monitoring guide documents what each major platform exposes; Shopify's online store, for example, offers no web server or edge log access, so merchants there cannot see crawler activity directly.

In the Comscore desktop panel, ChatGPT produced an outbound click in 5.2% of conversation sessions, against 31.1% of Google queries. Shi et al., Bocconi University, arXiv 2607.07652, 8 July 2026 Ecommerce Guide reading: taken together, these mean most referral figures on this page probably understate AI influence rather than overstate it, because the research step that happens inside an assistant and never produces a click is invisible to every referral dataset cited here. The only measurement that captures automated activity with verified identity is server-side, matched against published IP ranges and confirmed with reverse DNS, which are the two methods Google documents.

Forecasts, grade D, not measured results

Every statistic on this page links to the organization that produced the data, not to a report about it. Where a figure originates in a vendor's own release, earnings material, or a research paper, we link to that document. Where we have been unable to retrieve the underlying document, the figure has been removed rather than retained on trust. Each statistic carries its geography, the period measured, the sample or dataset size where disclosed, and the method used. Where two credible sources disagree, both are shown with their definitions, and we explain why they can differ rather than averaging them.

Published April 19, 2026. Last reviewed November 9, 2026, when every source below was opened and each figure checked against it. Any calculation performed by us is labeled "Ecommerce Guide calculation" or "Ecommerce Guide reading" in place. Maintained by the Ecommerce Guide editorial team; contact details are in the site footer.

Figures checked and not used

• A 4% merchant fee for ACP or Instant Checkout. No OpenAI, Stripe or Shopify document states a rate; OpenAI's merchant page says there are no fees on purchases that start in ChatGPT.
• "About 7 times" AI traffic growth for Shopify. It appears in third-party call transcripts, not in Shopify's releases, decks or investor site.
• $14.2 billion global and about $3 billion US AI-influenced Black Friday 2025 sales. No Salesforce-owned page states them or defines them.
• 805% AI traffic growth on Black Friday 2025. Adobe's own chart data labels 805% as season to date for 1 to 28 November 2025, not a single day, and no Adobe release states it. The season figure on this page is 693.4%.
• A count of merchants live on Instant Checkout. OpenAI has not published one.

Evidence grades

Grades describe the type of evidence, not the accuracy of the number. An A-grade panel can still be unrepresentative, and a C-grade company disclosure can be exactly right about the one company it describes.

A

Large first-party behavioral or transaction dataset: the Adobe Analytics panel, the Salesforce Shopping Index, Contentsquare's benchmark, the Kaiser and Schulze site panels, the Comscore household panel.

B

Credible survey with disclosed methodology, sample and fielding period.

C

Company disclosure, earnings material, product documentation, a laboratory benchmark, or a vendor dataset whose sample or period is not disclosed.

D

Secondary report, unclear methodology, or a directional estimate or forecast.