Be the store the AI shopping answer actually names

The market shift, and how fast it is moving

US retail e-commerce sales reached $326.7 billion in Q1 2026, up 9.8 percent year over year, well above the 3.9 percent growth of total retail, and accounted for 16.9 percent of all US retail sales that quarter, per the US Census Bureau's Quarterly Retail E-Commerce Sales report. That growth is not evenly distributed across discovery channels. Adobe Digital Insights, drawing on its first-party analytics panel across thousands of US retail sites, reports AI-referred traffic to US retail sites up 393 percent year over year in Q1 2026, and, critically, that in March 2026 AI-referred traffic converted 42 percent better than non-AI traffic, a reversal from March 2025, when it converted 38 percent worse. AI-referred shoppers also generated 37 percent more revenue per visit and spent 48 percent longer on-site.

Morgan Stanley projects agentic commerce could capture 10 to 20 percent of US e-commerce spend, $190 to $385 billion, by 2030. And the clearest evidence that the bottleneck is data, not shopper appetite, comes from OpenAI itself: when it relaunched shopping inside ChatGPT in February 2026, extending eligibility to over 1 million Shopify merchants against a base of roughly 900 million weekly active users, it pulled back from letting shoppers complete purchases inside the chat, citing inaccurate product data and difficult merchant onboarding as the specific reasons, not weak demand.

How today's shopper actually decides

Independent surveys converge on a real, if imprecise, finding: a majority of US shoppers begin product search on Amazon rather than Google, with estimates ranging from roughly 55 to 74 percent depending on the survey house. Reviews remain the dominant trust signal once a shopper narrows to a few options, with aggregated review-behavior research reporting over 91 percent of shoppers reading at least one review before buying, and purchase likelihood widely reported to peak in the 4.25 to 4.99 star range rather than at a suspiciously perfect 5.0.

Once a shopper has decided to buy, checkout is where the largest, best-documented loss happens. The Baymard Institute, from a decade of large-scale checkout testing spanning 50 pooled studies, reports an average cart abandonment rate of 70.22 percent, and that the typical large e-commerce site could gain roughly 35.26 percent more completed orders through better checkout design alone. Speed and mobile performance gate whether any of this matters: mobile now accounts for roughly 62 percent of e-commerce traffic, but only about 42 percent of mobile sites currently pass all three Core Web Vitals thresholds, versus 63 percent on desktop.

Who is actually running these stores

The businesses behind these numbers are, in the median case, small. Aggregated platform estimates put the average Shopify store's annual revenue at $72,000 to $235,000 depending on methodology, but the median store at only about $24,000 a year, meaning roughly half of all Shopify stores earn under $2,000 a month, while an estimated top 1 percent of stores generate roughly a fifth of all platform revenue. These figures are aggregator-sourced rather than platform-disclosed and should be read as directional, but the pattern they point to is consistent: the typical online merchant has no in-house data or SEO team to catch a feed disapproval, a schema mismatch, or a slow product page before it costs a sale.

The rules of agentic commerce are still being written

OpenAI's own pivot away from in-chat checkout in March 2026 is direct evidence that the standards for "buyable by an agent," the Agentic Commerce Protocol, the Universal Commerce Protocol, and platform-specific programs like Shopify's, are actively in flux rather than settled. That is not a reason to wait. It is the opposite: stores that get their product data, schema and feed clean now are positioned to qualify automatically as those standards solidify, rather than needing an emergency retrofit once they do.