TL;DR
- Our March 18, 2026 extract counted five stored technology signals. The six distribution rows contain 305,990 stores, with an average index of 2.03 out of 5.
- 29,836 stores (9.8%) had four or five signals; 28,371 (9.3%) had zero. Neither result measures whether products appear in AI shopping results.
- The index combines a theme-name rule, a detected SEO category, a Merchant Center tag, selected Google tags and app count. It does not inspect product data, schema validity, Catalog eligibility or an active feed.
- Current rules checked October 10, 2026: eligible ChatGPT discovery sends shoppers to the merchant's store checkout and has no selling fee beyond standard payment processing. Google, Copilot and Meta can offer direct checkout, subject to channel-specific requirements.
- Use the historical index to choose questions to investigate. A low score does not establish absent functionality, an agency budget or demand for an AI service.
Some links in this article are affiliate links. We may earn a commission if you purchase through them, at no extra cost to you.
Shopify's agentic storefronts let customers discover products through AI channels. Discovery, eligibility and checkout are separate questions. Our research describes visible technology signals across stored stores; Shopify's current product and channel controls answer the eligibility question more directly.
What We Actually Measured on March 18
We assigned one point to each of the following historical conditions, with equal weights. The index is exploratory and was not calibrated against AI placement, sales or merchant demand.
| Historical condition | Recorded rate | Interpretation and limit |
|---|---|---|
| Filled-in theme name outside the query's ten-name vintage list | 78.0% | A negative name-list match, not confirmation of OS 2.0, current version or AI compatibility |
| At least one detected app in the SEO category in the latest available snapshot | 1.7% | A category signature, not an inspection of JSON-LD or proof of missing native/theme/custom schema |
| Stored pixel name includes Google Merchant Center | 26.9% | A tag label, not Merchant Center approval or an active, accurate product feed |
| Stored pixel names include GA4, Universal Analytics, GTM or Google Ads | 70.1% | A historical tag union, including retired Universal Analytics; not current analytics operation or ad spend |
| Stored detected app count is at least three | 26.3% | A count of detected signatures, not paid subscriptions, activity or readiness |
The ten-name vintage list was Debut, Brooklyn, Minimal, Venture, Supply, Simple, Narrative, Boundless, Express and Launchpad. It was incomplete as an architecture classifier. Unknown or custom theme labels outside that list can earn the theme point. Vintage themes may still contain structured data; theme name alone does not establish whether Catalog discovery works.
Method: We retain the March 18 queries and output. The SEO condition uses the latest available snapshot without a maximum age; theme, pixels, app count, traffic tier and Plus classification come from stored fields. These are database records, not a confirmed panel of live merchants. Country eligibility was not applied.
Denominators: the independently executed overview reported 305,989 stores; the distribution and displayed traffic rows each sum to 305,990; the final summary and Plus split report 305,991. We preserve those scopes rather than pretending the extract was one frozen read of the data. The distribution uses its actual row sum below. Snapshot-age distribution was not retained, so the study cannot establish how current each observation was. No query was rerun for this revision.
The Historical Signal Distribution
| Signals counted | Stores | Share of the 305,990 distribution rows |
|---|---|---|
| 0 | 28,371 | 9.3% |
| 1 | 76,191 | 24.9% |
| 2 | 90,612 | 29.6% |
| 3 | 80,980 | 26.5% |
| 4 | 28,416 | 9.3% |
| 5 | 1,420 | 0.5% |
The rounded percentages do not sum exactly to 100%. 195,174 stores (63.8%) had at most two signals; 29,836 (9.8%) had at least four. Zero means none of these five conditions matched. It does not mean no product feed, no schema, or invisibility to an agent.
Traffic Groups
| Estimated monthly traffic tier | Stores | Mean index | Four or five signals | Zero signals |
|---|---|---|---|---|
| Under 50K | 257,077 | 1.81 | 5.0% | 11.0% |
| 50K–200K | 46,887 | 3.17 | 35.0% | 0.2% |
| 200K–1M | 1,984 | 3.15 | 28.8% | 0.0% |
| 1M–5M | 35 | 2.97 | 5.7% | 0.0% |
| 5M–20M | 7 | 3.00 | 0.0% | 0.0% |
The 50K–200K group had a higher mean index than the under-50K group. This is a cross-sectional association with estimated traffic, not evidence that adding tags improves discovery or that merchants have a budget. The two largest traffic ranges contain only 35 and seven stores; they do not support a general conclusion about large-store readiness or which merchants will earn AI revenue first.
Categories
| Stored category | Stores | Mean index | Four or five signals | Zero signals |
|---|---|---|---|---|
| Beauty | 21,142 | 2.26 | 13.9% | 7.0% |
| Baby & Kids | 6,588 | 2.25 | 12.8% | 6.6% |
| Health & Wellness | 15,419 | 2.22 | 13.0% | 7.1% |
| Pets | 5,350 | 2.17 | 12.9% | 7.8% |
| Electronics | 8,500 | 2.11 | 8.3% | 6.5% |
| Food & Beverage | 27,542 | 2.10 | 11.1% | 8.4% |
| Home & Garden | 43,546 | 2.05 | 9.7% | 8.7% |
| Fashion | 70,493 | 2.01 | 9.5% | 9.5% |
| Jewelry | 20,220 | 2.00 | 9.0% | 10.0% |
| Hobby | 27,139 | 1.83 | 7.0% | 12.0% |
These are ten selected rows from the saved category output, not an exhaustive ranking. Beauty had the highest recorded category mean in that output. A category's lower index does not show less capable founders, more solo operators or an addressable service market. Fashion's 70,493 records describe the cohort size, not 64,000 merchants who need paid work.
Stored Plus Classification
| Segment in its separate 305,991-store query | Stores | Mean index | Four or five signals | Zero signals |
|---|---|---|---|---|
| Classified Plus | 53,492 | 3.35 | 42.2% | 0.0% |
| Standard | 252,499 | 1.75 | 2.9% | 11.2% |
The index is higher in records classified Plus. Neither plan classification nor technology count establishes the team size, purchasing intent or likely AI revenue. It also does not define channel eligibility.
The separate theme-era query recorded 67,365 names in the ten-name vintage list (22.0%), alongside 101,363 paid-theme labels, 69,935 modern-free labels, 56,777 custom labels and 10,549 other/unknown labels, totaling 305,989. Those broad labels are not version checks or measured compatibility defects. Do not sell an urgent migration solely because a name appeared in this group.
How Agentic Channels Work Today
The following summarizes official Shopify help checked October 10, 2026, after the historical study. Controls are under Sales channels > Agentic. Eligible defaults and availability differ by channel; inspect the merchant's actual admin and product requirements rather than assuming every Shopify store is enrolled. See Shopify's overview.
| Channel | Eligibility or discovery route | Checkout and selling fee |
|---|---|---|
| ChatGPT | Sell to US customers; the merchant can be outside the US. Catalog-eligible products, supplemental terms and completed Terms, Privacy and Return/refund policies are required | Store's online checkout in an in-app browser or new web tab. No ChatGPT selling fee; standard processing fees apply |
| Google AI Mode and Gemini | Direct checkout requires a US-based store selling to US customers, valid Merchant Center account/products, Catalog eligibility and required policies/terms | Direct checkout is rolling out and may be unavailable; otherwise redirect to the store. No direct-checkout selling fee beyond standard processing |
| Microsoft Copilot | Sell to US customers, with eligible Catalog products and required policies/terms | Supported direct checkout displays to US customers; can be deactivated for redirect. No selling fee beyond standard processing |
| Meta | Sell to US, Canadian or Mexican customers, with eligible Catalog products and required policies/terms. Shoppable ads require Facebook and Instagram product syncing | Supported direct checkout displays in those markets. No selling fee beyond standard processing; checkout blocks and several product types are unsupported |
Google discovery uses Google Merchant Center, through Google & YouTube, a feed provider or manual upload. A Merchant Center tag in our historical index does not validate that account or feed. Shopify Catalog supplies discovery to other supported channels through a different route; Google feed state cannot be inferred from Catalog access.
For ChatGPT, eligible products can be discovered through Catalog without installing an SEO app. Catalog access does not guarantee a recommendation for a particular prompt. Google/Copilot direct checkout also has feature restrictions: subscriptions, bundles and customizable products are not supported there, and standard client-side analytics/custom pixels do not fire. Check requirements before assuming the store's existing checkout experience transfers intact.
To manage individual settings, review Shopify's channel controls. Disabling Shopify Catalog access may take up to seven days to stop data sharing and does not prevent all discovery through web crawling. ChatGPT has no direct-checkout toggle. Merchants should choose channel settings based on products, policy and workflow requirements.
OpenAI's March 24, 2026 product-discovery announcement describes a focus on discovery and merchant checkout flexibility.
A Practical Merchant Review
The historical index supplies questions, not an automated diagnosis. A proposed review should produce evidence the merchant can inspect:
| Question | Concrete check | What to record |
|---|---|---|
| Are these products eligible? | Review current Catalog and each channel's rules with the merchant | Product, rule, channel and date; distinguish prohibited/B2B-only products from data errors |
| Is the product data accurate? | Compare title, description, price, stock, variants, materials and policies with the actual offer | Specific mismatches and a correction owner, rather than a generic completeness score |
| Is the appropriate discovery route working? | With authorized admin access, inspect Catalog preview or Merchant Center product/feed diagnostics | Status, errors and check time; a public tag is insufficient |
| Does the intended journey work? | Plan merchant-approved channel and checkout tests for relevant variants, discounts and required inputs | Actual task and result, including channel restrictions; do not invent an order test |
| Is the result measurable? | Review available channel/referrer reporting and compare like periods | Sessions, attributed orders and reporting limitations; discovery and revenue are separate outcomes |
Shopify's Catalog guide and merchant preview are a better starting point than an installed-app count. Inspect actual structured data if that is part of the agreed task; native, theme or custom implementation may already supply it.
If a merchant needs clearer policies or AI-facing answers, review Shopify's own tools before proposing a new app. Scope the work to verified data issues and required workflows. We have not measured market rates, delivery margin or demand for any service package, and no universal schema lift or citation-decay interval is established by this study.
For an app idea, interview merchants and check current native and competing solutions before building. The number of stores without a tracked SEO-category app is not a schema-tool TAM. Use the app-idea validation guide for a separate demand test.
Frequently Asked Questions
What percentage of Shopify stores are AI-ready?
This study cannot answer that. In the March distribution, 9.8% had four or five selected signals, but the equal-weight index was not validated against product eligibility, actual discovery or sales.
Must a store install an SEO app or modern theme to appear in ChatGPT shopping?
The historical signals do not establish such a requirement. Shopify's current help describes automatic Catalog discovery for eligible products. Check product/channel eligibility and the store's actual implementation before proposing an app or migration.
What does selling through ChatGPT cost?
Shopify's October 10 help check states no associated ChatGPT selling fee, with standard payment processing fees applying. Check the relevant channel's current terms before budgeting.
How can an agency use the study responsibly?
Use the cohort tables to select a small research sample, then inspect product data and channel state with permission. Confirm a real problem and the merchant's interest before pitching. Pixels, traffic estimates and low index values do not prove active spend, budget or intent.
Find Shopify Clients Worth Your Time
Search by niche, traffic, and tech stack. Export with verified founder contacts.Search stores by niche, traffic, and tech stack. Export with verified founder contacts so you can skip the research.
![Shopify Shipping Protection Apps [136,170-Store Study]](/_next/image?url=%2Fimages%2Fblog%2Fshopify-shipping-protection-apps.webp&w=3840&q=75&dpl=dpl_6TQooVZF25SXcyYFNhA2QiNSjjgH)
![Shopify Sticky Add to Cart [133,357-Store Study]](/_next/image?url=%2Fimages%2Fblog%2Fshopify-sticky-add-to-cart.webp&w=3840&q=75&dpl=dpl_6TQooVZF25SXcyYFNhA2QiNSjjgH)
![How Accurate Is Shopify App Detection? [50-Store Test]](/_next/image?url=%2Fimages%2Fblog%2Fshopify-app-detection-accuracy.webp&w=3840&q=75&dpl=dpl_6TQooVZF25SXcyYFNhA2QiNSjjgH)