Agentic Commerce Statistics 2026: What the Data Says About AI Shopping

What does the data say about agentic commerce in 2026?

The measurable part of agentic commerce in 2026 is discovery, not checkout. More shoppers click from AI assistants to retail sites, and Adobe's data shows those visits now convert better than other traffic. Buying inside the chat window has a weaker record. OpenAI scaled back its native Instant Checkout in March 2026, after Walmart reported that in-chat conversion was three times lower than click-out conversion. Google is still building checkout into its AI surfaces, but its new cart also lets shoppers finish on the merchant's site.

The traffic is also still small. In Semrush's sample of 19 large retailers, AI referrals were about 0.15% of all visits over the 2025 holidays.

Disclosure: This article is published by CiteLadder, which sells software that tracks how brands appear in AI answers. Every figure comes from a third party. We opened and read each source on 20 October 2026; the date next to a figure is the publisher's date or data period. Sections labeled "Our reading" are our interpretation, not data.

Key agentic commerce statistics at a glance

What is agentic commerce?

Google describes agentic commerce as AI completing tasks on people's behalf.[13] McKinsey describes agents that navigate options, negotiate deals and execute transactions.[17] Forrester points out that the market uses the term loosely, and that today's answer-engine shopping has less autonomy than the word implies.[8]

To read the data, split it into two layers:

• Discovery. An assistant answers a shopping question, names products and links out. The purchase happens on the merchant's site. This layer is measurable today as referral traffic and as appearances in AI answers.
• Transaction. An assistant or agent completes checkout inside the AI surface, using a protocol such as OpenAI's Agentic Commerce Protocol (ACP) or Google's Universal Commerce Protocol (UCP).

Most public numbers describe the first layer. The second layer has launch posts, partner lists and a few retailer statements, but almost no published outcome data.

Discovery: AI referrals are growing fast from a small base

Adobe: steep growth rates

Adobe Analytics covers more than 1 trillion visits to US retail sites, according to Adobe. For the 2025 holiday season, Adobe reported a 693.4% year-over-year increase in traffic from generative AI tools, counted as shoppers clicking a link to a retail site.[1] Adobe added that the base of users "remains modest."[1]

Later Adobe releases show lower growth rates:

• Retail AI visit share up 393% year over year in Q1 2026.[3]
• AI-driven traffic to retail sites up 127% year over year in August 2026.[5]
• A forecast of 130% growth for the 2026 holiday season.[4]

These are not identical metrics. The holiday 2025 figure is traffic growth; the Q1 2026 figure is growth in AI's share of visits. Adobe's own 2026 forecast pages also differ on one detail: the press release expects AI traffic on Thanksgiving to rise 159%,[4] while the report page says 141%.[5]

Semrush: a small share of total traffic

Semrush's data shows the size of the channel. Using Semrush Traffic Analytics for 19 large retailers from November 1, 2025 to January 9, 2026, it counted 5.87 billion total visits, 1.055 billion from organic search and 8.87 million from AI referrals.[6] That puts AI at roughly 0.15% of traffic and organic search at roughly 18%, or about 119 times more visits from search.[6]

AI's share reached about 0.216% on December 31, one of the highest daily shares in the dataset. Retailers in higher-consideration categories had above-average AI shares, including Best Buy at about 0.30% and Home Depot at about 0.24%.[6]

Why the two pictures fit together

Adobe reports growth rates from first-party analytics on its clients' sites. Semrush reports a share of visits from third-party traffic estimates for a fixed set of large retailers. A channel can grow several times over and still be a fraction of a percent of visits. Neither source, in the pages we read, explains how it treats AI app visits that arrive without a referrer.

Do AI-referred shoppers convert?

Adobe says yes, and the change is recent. During the 2025 holidays, AI referrals converted 31% more than other traffic sources.[2] Adobe also reported AI-driven revenue per visit up 254%, which it described as "this holiday season year-to-date."[2]

The Q1 2026 report shows the reversal more clearly. In March 2026, AI-referred retail visitors converted 42% more than non-AI visitors. A year earlier, they converted at nearly half the non-AI rate.[3] Revenue per AI visit was 37% higher in March 2026; twelve months before, non-AI visits were worth 128% more.[3]

Walmart adds a second signal. Daniel Danker, a Walmart executive vice president, told WIRED that ChatGPT brings in about twice the rate of new customers that search engines do.[12]

Two limits apply. Adobe's figures cover US sites that use Adobe Analytics. They also measure what happens after a click to the retailer's own site, so they say nothing about sales completed inside an AI tool.

Checkout inside the chat: the record so far

OpenAI Instant Checkout, September 2025 to March 2026

OpenAI launched Instant Checkout on September 29, 2025, together with the Agentic Commerce Protocol, built with Stripe.[9] US users on the Free, Plus and Pro plans could buy single items from US Etsy sellers in chat, and OpenAI said over a million Shopify merchants were coming.[9] Merchants paid a small fee on completed purchases. OpenAI said Instant Checkout items were not preferred in product results, though checkout availability was one factor when ranking merchants that sell the same product.[9]

Uptake stayed small. Forrester analyst Emily Pfeiffer told CNBC that roughly 30 Shopify merchants were available through Instant Checkout as of February 2026. Walmart confirmed it had made about 200,000 products available.[11] Danker told WIRED that conversion for items sold directly inside ChatGPT was three times lower than for items that required clicking out to Walmart's site.[12]

In March 2026, OpenAI confirmed it was ending Instant Checkout in favor of retailer apps that send shoppers to the retailer's own site to pay.[11] On March 24, 2026, OpenAI explained the change itself. It said the first version did not give merchants enough flexibility, so merchants can now use their own checkout while OpenAI focuses on product discovery.[10]

The same post extended ACP to carry merchant product feeds and promotions. It named Target, Sephora, Nordstrom, Lowe's, Best Buy, The Home Depot and Wayfair as integrated for discovery. Shopify merchants' catalogs flow in through Shopify Catalog, and purchases finish on the merchant's store in an in-app browser.[10] Walmart launched an app inside ChatGPT that supports account linking, loyalty and Walmart payments.[10]

Google: UCP, then a cross-retailer cart

Google launched UCP on January 11, 2026. It was co-developed with Shopify, Etsy, Wayfair, Target and Walmart.[13] Google said UCP would power checkout from eligible US retailers in AI Mode and the Gemini app through Google Pay, with the retailer as seller of record. The same post introduced Business Agent, a branded chat on Search, and Direct Offers, a Google Ads pilot that shows exclusive discounts in AI Mode.[13]

In March 2026, Google added optional cart, real-time catalog and identity-linking capabilities to UCP.[14] On September 16, 2026, Google said UCP enabled direct checkout for "hundreds of thousands of brands and retailers," and added cart transfer to merchant sites.[15] On October 15, 2026, Universal Cart began rolling out in the US across Search and the Gemini app. Shoppers can check out with Google Pay or move items to the merchant's own cart.[16]

The Google posts we read report no conversion or adoption figures for in-surface checkout.

What shoppers say about buying inside AI tools

The surveys seem to disagree, mostly because they ask different people different questions.

• Semrush surveyed 1,030 US shoppers who had tried AI tools, in December 2025. 50% said they had bought something after using AI for research, and 22% said they had completed a purchase inside an AI tool.[7] Semrush sells AI search tools, and its page does not define "inside an AI tool."
• Forrester found that completing a purchase inside an answer engine was the least-adopted use case among regular answer-engine users in the US, UK and Canada, in a March 2026 survey.[8]
• Adobe's March 2026 survey of more than 5,000 US consumers found that 39% had used AI assistants for online shopping. Half of consumers said they click links an assistant provides when shopping with AI, and 27% said they buy through those links.[3]

Our reading: Semrush asked AI users whether they had ever bought inside an AI tool. Forrester ranked use cases. Adobe asked about buying through links, which ends on the merchant's site. All three can be true at once. None of them shows buying inside the chat as a common habit.

Projections are not measurements

McKinsey projects that by 2030, agentic commerce could orchestrate $900 billion to $1 trillion of US B2C retail revenue, and $3 trillion to $5 trillion globally.[17] McKinsey says the estimate assumes moderate merchant readiness, covers goods only, and excludes services and B2B. The page we read does not define "orchestrated revenue." Treat it as a scenario, not a forecast of sales completed inside AI tools.

Many 2026 statistics roundups repeat other large figures, including a Salesforce estimate of holiday revenue driven by AI and agents. We could not open a primary source for those figures, so we left them out.

Our reading: what this means for product visibility

This section is CiteLadder's interpretation of the data above.

1. Discovery is where the value is measurable now. Adobe's conversion data and Walmart's new-customer comment both describe shoppers who clicked out. Being named and linked in the answer is what produces that visit. That is a question of AI visibility, more than of checkout integration.
2. Checkout is moving back toward the merchant. OpenAI's March 2026 change and Google's cart transfer both let the merchant complete the sale. Forrester's advice to merchants points the same way: focus on discoverability and keep the checkout moment.[8]
3. Product data is what both platforms ask for. OpenAI now takes product feeds through ACP.[10] Google asks for conversational attributes in Merchant Center. It reports that in a test with lululemon, brand-submitted attributes were used 50% of the time in relevant AI Mode recommendations, and that merchants following its feed best practices see on average a 5% conversion increase the next month.[15] Both are Google-reported figures.
4. Platforms now report share of voice themselves. Google's AI performance insights in Merchant Center compares a brand's share of voice with similar brands across AI Mode and AI Overviews, in five countries.[15] It covers Google surfaces only.
5. Page quality and AI traffic move together in Adobe's sample. Retailers in the top 20% by AI visit share scored higher on Adobe's "AI citation readability" measure, with the largest gap on homepages, at 52%.[3] That is a correlation in Adobe's own scoring, not proof of cause.

How to measure AI shelf share for your products

1. Write buyer-style product and category questions for each market, such as "quietest cordless vacuum for a small apartment."
2. Run them on a schedule across the engines your shoppers use, and keep the raw answers.
3. Record which products and brands appear, in what order, and which sources are cited. Track AI citations separately from product mentions.
4. Compare your appearance rate with named competitors on the same questions.
5. Segment AI referral traffic in your analytics. Compare its landing pages and conversion with other channels. Some AI app visits may arrive without a referrer and be counted as direct.
6. Change one input at a time, such as feed attributes, and judge the result over repeated runs, not single answers.

CiteLadder tracks how brands appear in ChatGPT, Claude, Gemini and Google AI Overviews answers. For eligible projects, Commerce Intelligence measures product and category appearances in AI answers, alongside analysis of the cited sources. The code is open source under AGPL-3.0, and a hosted version runs at citeladder.com.

Frequently asked questions

What is agentic commerce?

It is shopping where AI acts for the buyer: finding products, comparing them and, in some cases, completing the purchase. In 2026, most measurable activity is the discovery step. The assistant recommends products and the shopper buys on the merchant's site.

How much traffic does AI send to retail sites?

Not much yet. Semrush estimated AI referrals at about 0.15% of visits to 19 large retailers over the 2025 holidays, against about 18% from organic search.[6] Adobe reports fast growth: 127% year over year in August 2026.[5]

Can shoppers still buy inside ChatGPT?

OpenAI moved away from native Instant Checkout in March 2026 and lets merchants use their own checkout.[10] Merchants can build ChatGPT apps. Walmart's app, for example, supports Walmart payments inside ChatGPT. Shopify merchants' sales finish on the merchant's own store in an in-app browser.[10]

Do shoppers from AI assistants convert better?

In Adobe's US retail data, yes: 42% higher conversion and 37% higher revenue per visit than non-AI traffic in March 2026.[3] A year earlier, AI visitors converted at about half the non-AI rate. Your own site may differ, so measure it.

Is the "$1 trillion agentic commerce" figure real?

It is a McKinsey projection for US B2C retail by 2030, based on moderate assumptions about merchant readiness.[17] It is not a measurement of current sales.

Sources

All sources opened and read on 20 October 2026.

1. Adobe, Holiday Shopping Season Drove a Record $257.8 Billion Online with Consumers Embracing Generative AI Tools, 7 January 2026.
2. Adobe, AI-driven traffic surges across industries with retail experiencing biggest gains, January 2026.
3. Adobe Digital Insights, Quarterly AI Traffic Report, April 2026.
4. Adobe, U.S. Holiday Shopping Season to Hit Record $275.1 Billion Online, 28 September 2026.
5. Adobe, 2026 Holiday Shopping Trends, Statistics and Forecast, 28 September 2026.
6. Semrush, Holiday Shopping 2025: What Traffic Data Reveals About Search, AI, and How People Actually Buy, data period 1 November 2025 to 9 January 2026.
7. Semrush, How AI Tools Influence the Modern Buyer Journey: A Survey of 1,000+ US Consumers, survey fielded December 2025.
8. Forrester, What It Means That The Leader In "Agentic Commerce" Just Pulled Back, March 2026.
9. OpenAI, Buy it in ChatGPT: Instant Checkout and the Agentic Commerce Protocol, 29 September 2025.
10. OpenAI, Powering Product Discovery in ChatGPT, 24 March 2026.
11. CNBC, OpenAI's first try at agentic shopping stumbled. It's trying again, 20 March 2026.
12. WIRED, Why Walmart and OpenAI Are Shaking Up Their Agentic Shopping Deal, March 2026.
13. Google, New tech and tools for retailers to succeed in an agentic shopping era, 11 January 2026.
14. Google, AI shopping gets simpler with Universal Commerce Protocol updates, 19 March 2026.
15. Google, Boost your holiday sales with these agentic commerce updates, 16 September 2026.
16. Google, Google's new Universal Cart launches today in the U.S., 15 October 2026.
17. McKinsey, The agentic commerce opportunity: How AI agents are ushering in a new era for consumers and merchants, report page shows no date.