Shopify Says AI Search Traffic Tripled in Q2, and 75% of Those Sales Came From Outside Its Biggest Categories. Here's What That Means for Smaller Sellers

A lot of the anxiety around AI search has centered on a real, documented pattern, AI summaries pulling traffic away from the sites they summarize, the way they have for online publishers. Shopify's second-quarter results tell a different story for e-commerce, and the specific reason why is worth understanding closely.

The Problem: Sellers Have Real Reason to Worry About AI Search Cannibalizing Traffic

The fear isn't imaginary. AI-generated search summaries have measurably reduced click-through rates for online publishers, since a shopper or reader getting a complete answer directly in an AI response has less reason to click through to the original source. It was reasonable to assume ecommerce would follow the same pattern, AI answering a shopping question directly instead of sending a shopper to a merchant's actual product page.

What Shopify Actually Reported

On its Q2 earnings call, Shopify President Harley Finkelstein described AI as a "complement to search, rather than a substitute for it." The company's actual numbers back that framing up directly, not just as a talking point.

• AI-driven traffic and orders to Shopify stores tripled year over year in the second quarter.
• Traditional search traffic didn't shrink to make room for it. Search sessions are up 1.3x over the past two years and still hold roughly a third of all storefront sessions.
• Revenue rose 36% to $3.6 billion, ahead of Wall Street's $3.4 billion forecast, and gross operating profit rose 31% to $1.71 billion, also ahead of expectations.
• Half of all AI-referred sessions landed directly on a product description page, 2.5 times the rate for traditional search referrals, meaning AI-driven traffic skips more of the browsing funnel and arrives closer to a purchase decision already formed.
• 75% of AI-attributed purchases in Q2 happened outside Shopify's top 100 categories, what the company calls its "sweet spot," a genuinely striking number worth sitting with.

Why This Is Different From What's Happening to Publishers

Finkelstein's own explanation of the mechanism is the actual useful part here, not just the headline numbers. Traditional search ranks by popularity against a handful of keywords. AI agents instead make multiple calls into a merchant's catalog, working with richer structured product data to match a buyer's specific intent rather than just keyword overlap.

His example: a shopper asking an AI assistant for the best car seat that fits three across in a sedan. Traditional search optimizes around the keyword "car seat." An AI agent parses the actual constraints, the dimensions, the vehicle type, the number of seats needed, and searches across all of them simultaneously to find a product that genuinely fits, not just the listing that ranks highest for a generic term.

What This Means for Smaller Sellers Specifically

The 75%-outside-top-100-categories figure is the number worth paying the most attention to if you're not already dominating a major category. Traditional search ranking rewards scale and popularity, which structurally favors large, established players in competitive categories. AI agents querying across specific product attributes to match a precise buyer need don't automatically favor the biggest catalog or the most reviews, they favor whichever product's actual data genuinely matches what the buyer asked for. A niche product with accurate, detailed structured data has a real shot at getting surfaced for a highly specific query, even without the search ranking history a larger competitor has built up over years.

This also means the traditional SEO playbook, optimizing for keyword density and backlink authority, doesn't fully transfer to AI-referred traffic. What matters more is whether your product's actual data, dimensions, materials, compatibility, use case specifics, is complete, accurate, and structured in a way an AI agent can actually query against.

The Broader Context Worth Knowing

This divergence from publisher traffic loss isn't unique to Shopify's specific numbers, it reflects a broader pattern industry analysts are tracking across ecommerce generally. Marketplace Pulse's ongoing coverage of ecommerce platform trends tracks how different platforms are adapting their catalog data and structured content specifically to be more queryable by AI agents, a trend worth watching beyond Shopify alone. McKinsey's framework for how AI agents are starting to shop and transact on a buyer's behalf also covers the broader shift toward intent-based, constraint-matching discovery replacing pure keyword ranking, which is the exact mechanism Finkelstein described on the earnings call.

What to Actually Do About This

1. Audit your product data completeness, not just your product descriptions. Dimensions, materials, compatibility details, and specific use-case attributes are exactly what an AI agent queries across. A product page with a compelling description but thin structured data is optimized for the wrong kind of search.
2. Prioritize your product description pages, not just collection or homepage content. Since half of AI-referred sessions land directly on a PDP, that page needs to independently answer a buyer's specific question without relying on surrounding site navigation to provide context.
3. Think in terms of specific buyer constraints, not broad keywords. The car seat example is the model: what are the two or three specific, factual details that would let an AI agent confidently match your product to a precise need, and are those details actually present and accurate on your listing?
4. Don't assume this only helps merchants in Shopify's biggest categories. The data suggests the opposite, a well-documented niche product may have a real structural advantage in AI-referred discovery that it never had in traditional keyword-ranked search.

Shopify has also built connectors to Claude, ChatGPT, Perplexity, Manus, Replit, Vercel, and Lovable, meaning AI tools and coding platforms can already interact with a merchant's store directly, worth knowing if you're evaluating how AI-assisted shopping tools might interact with your specific storefront going forward. Sellers already using Shopify's AI-connector ecosystem, or considering it, are in a reasonable position to benefit from this shift faster than those treating structured data as an afterthought.

Where This Connects to Your Existing Optimization Work

If you've already invested in Shopify Magic for product descriptions, our guide on using AI in Shopify Magic is worth revisiting with this new context, description quality and structured data completeness are related but distinct goals, and this data suggests the structured side deserves at least equal priority. For the parallel story on Amazon, our coverage of breaking into Rufus and AI search results covers the same underlying shift on a different platform.

For the original reporting this post is based on, see TechCrunch's coverage of Shopify's Q2 earnings call.

Frequently Asked Questions

Is this data specific to Shopify, or does it apply to all ecommerce platforms?

The specific numbers are Shopify's own reported figures from its Q2 earnings call. The underlying mechanism, AI agents querying structured product data across multiple buyer constraints rather than ranking by keyword popularity, is a platform-general pattern, though actual traffic results will vary by platform and how well-structured a given catalog's data actually is.

Does this mean traditional SEO no longer matters for Shopify stores?

No, traditional search sessions are still up and still represent roughly a third of storefront traffic according to Shopify's own reporting. This is additive, not a replacement, worth treating AI-referred traffic as a new channel to optimize for alongside existing SEO work, not instead of it.

What specifically should I check in my product data first?

Start with dimensions, compatibility, materials, and any specific use-case attributes that distinguish your product from close alternatives, exactly the kind of detail an AI agent needs to match a buyer's specific constraints, as illustrated in Shopify's own car seat example.

Why did 75% of AI-attributed purchases happen outside Shopify's top 100 categories?

Shopify's own explanation centers on how AI agents match specific buyer intent against structured data rather than ranking by popularity, which doesn't automatically favor the largest or most reviewed products the way traditional keyword search does, giving well-documented niche products a more even footing.

Should I worry about AI summaries reducing my site's traffic the way they have for publishers?

Based on Shopify's Q2 data, this pattern hasn't played out the same way for ecommerce, AI-driven traffic and orders grew substantially alongside stable traditional search traffic, rather than cannibalizing it. The dynamics driving publisher traffic loss, AI providing a complete answer directly, don't map cleanly onto a shopping decision that still requires an actual purchase transaction.

Takeaways

• Shopify reported AI-driven traffic and orders to its stores tripled year over year in Q2, without traditional search traffic shrinking to make room for it.
• 75% of AI-attributed purchases happened outside Shopify's top 100 categories, a real structural opportunity for smaller and niche sellers.
• AI agents match buyer intent across structured product data and multiple constraints, unlike traditional search's keyword-and-popularity ranking, which is why the traffic pattern differs so much from what's happened to publishers.
• Half of AI-referred sessions land directly on a product description page, meaning that page needs to stand on its own without relying on site navigation for context.
• Auditing structured product data completeness, dimensions, materials, compatibility, specific use cases, is now a distinct optimization priority alongside traditional SEO and description quality.

For ongoing coverage of AI search and ecommerce platform trends, see our newsletter.