Shopify's Q2 2026 results point to a new way for merchants with specific products to be found: AI shopping tools that match products to what a buyer actually needs.
Shopify reported that AI-driven traffic and orders to its stores each tripled year over year, while traditional search sessions were also up, by 1.3x, though measured over two years. That is a different story from the one publishers tell about AI and lost clicks. The figures are Shopify's own, and Shopify itself said agentic commerce volume is still small relative to its GMV. Treat the results as directional.
What Shopify Reported
Shopify announced its Q2 2026 results (April to June) on August 5. Retail TouchPoints summarised the AI figures from the earnings call:
| Metric (Q2 2026) | What Shopify reported |
|---|---|
| AI-driven traffic to stores | Tripled year over year |
| AI-driven orders | Tripled year over year |
| New-buyer orders from AI channels | Nearly twice the rate of other channels |
| AI-attributed purchases outside the top 100 categories | 75% |
| AI-referred sessions landing on product pages | About half, 2.5x the rate for traditional search |
| Search sessions | Up 1.3x over two years; about one-third of storefront sessions |
Finkelstein cited a screen-less phone for children, reef-safe sunscreen and a dog harness for French bulldogs as products that surfaced through AI search: specialized goods of the kind most Shopify merchants sell.
Shopify's figures cover AI-driven traffic and orders and don't separate click-throughs from purchases completed by an AI agent. Catalog is Shopify's structured product-data layer for AI discovery; carts and checkout run through the Universal Commerce Protocol (UCP). Nothing in the coverage shows Catalog alone drove the growth.
Why this could favour smaller merchants
Finkelstein's argument is that AI shopping is "merit-based", resulting in AI agents that aren't driven by advertisers who paid more. He made the case in a January 2026 Retail Brew interview and again on TBPN, as relayed by The AI Daily Brief. As the show summarizes it, agents match a buyer's specific constraints, such as a car seat that fits three across in a sedan, instead of ranking by popularity.
Directionally, the data indicates that if AI agents match on product attributes, then a small store with complete, specific product data can compete for a buyer who has a narrow need. A large retailer's advantage in brand awareness and ad budget could count for less in some searches. It would not remove the influence of platform coverage, product availability, ranking or advertising.
What merchants can do now
These steps follow from Shopify's disclosures and from standard search practice. They are reasonable bets, not proven formulas.
Complete your product data. Fill in titles, descriptions, images, categories, variants, options, prices and inventory, as Shopify advises. Use product fields and metafields for attributes such as materials, dimensions, compatibility and intended use, and confirm they reach the catalog or feed the shopping tool uses. Start with Shopify's Catalog help page.
Answer shoppers' specific questions. Does it fit, who is it for, what is it made of. Write for real shoppers, not for AI.
Make each product page stand on its own. About half of AI-referred sessions land on product pages, so show price, availability, shipping, returns and reviews there.
Keep SEO fundamentals and feeds sound. Pages should be crawlable and indexable. Google says llms.txt files and special markup are unnecessary for Google Search. Keep Merchant Center feeds and Google Business Profiles current, and check eligibility and data mapping.
Earn genuine reviews and mentions. They help shoppers and may inform some AI experiences, but don't manufacture them.
Measure in three stages. Track visibility, referral traffic and orders. Shopify's admin has an agentic section with cross-channel attribution (added in May). In GA4, segment identifiable AI referrals, and check Search Console's generative AI report. Some AI visits carry no referral data, so expect undercounting.
What to keep in mind
Four caveats before you act on any of this:
The figures are Shopify's own, and the base is small. Shopify said agentic commerce volume is still small relative to its GMV, and the coverage gives multiples, not absolute counts. The 75% figure shows where AI-attributed purchases came from, not which merchants win them.
Catalog inclusion is no guarantee. Shopify's Help Center says inclusion doesn't guarantee a product appears in a given AI answer or ranks in any position, and each channel controls final ranking. See Shopify's Catalog help page.
Wider evidence is thin. A New York Fed analysis of U.S. employer firms found AI users more likely to expect higher revenue, but only 31% of respondents said AI had increased their sales. It covers AI adoption generally, not AI shopping.
GEO is not a shortcut around SEO. Google's 2026 guidance says optimizing for its generative AI features is still SEO, and advises against special AI markup or manufactured mentions. Other AI tools may work differently. Treat any promise of guaranteed citations with suspicion.
Want a second pair of eyes on your catalog?
I work with Shopify merchants to review how their product data and pages read to AI shopping tools, and set up tracking to see what AI search sends them. No one can promise AI citations, but a catalog audit shows where your product data is thin. To book a review, contact ne (Sandi) at snelson@smarketingstudio.com.