Benchmarks / All stores
Agentic commerce benchmark across all stores
Across 251 stores Ottom measured at every stage in the 90 days to 25 September 2026, in every category and including stores whose category could not be determined, the median agentic readiness score was 55 out of 100. 2% of them let an AI shopping agent reach the payment step.
This benchmark pools every store Ottom has scanned, across all categories, including stores whose category could not be determined. Each store is counted once, whichever category it is in. It is counted under exactly the rules each category is, set out in the methodology.
n = 251 of 254 stores · 27 June 2026 to 25 September 2026 · bench-v4 · methodology
Scores come from the 251 stores Ottom measured at every stage. A further 3 were scanned but stopped before checkout, usually at bot protection or a robots.txt rule, so they have no comparable score. All 254 are still counted in the stalls below, because what was found on a store does not depend on how it was scored.
A store enters Ottom’s survey only if a catalogue of at least ten products can be read from it in one request, which in practice means Shopify and WooCommerce storefronts. These figures therefore describe the storefronts Ottom can read in bulk rather than online retail as a whole. The methodology sets out what that excludes.
Readiness score
The median of the 251 stores Ottom measured at every stage in the 90 days to 25 September 2026 scored 55 out of 100, and a quarter of them scored 52 or below.
The spread of overall readiness scores across the category. We publish quartiles rather than a best and worst store, because the best and the worst are each one identifiable merchant.
Where agents stall
Of the 251 stores Ottom measured at every stage, 2% scored 80 or better at the buy stage, against 91% at the compare stage.
Each stage of the journey, and the share of stores whose score at that stage was 80 or better.
| Stage | Median score | Stores passing | Stores measured |
|---|---|---|---|
| Discover | 75 | 39% | 251 |
| Evaluate | 88 | 87% | 251 |
| Compare | 100 | 91% | 251 |
| Buy | 0 | 2% | 251 |
Reaching the payment step
2% of the 251 stores Ottom measured let an AI shopping agent reach the payment step. Ottom stops there and never submits payment.
What stops them
The most common stall across all stores is products have no GTIN or MPN to match against, found on 66% of the 254 stores Ottom scanned in the 90 days to 25 September 2026.
The most common stalls across all stores, as a share of all 254 stores scanned. A stall found on fewer than 5 stores is withheld, because at that count it points at a merchant rather than at a pattern.
- No sitemap57%
- No product structured data9%
- Bot wall6%
- No product identifier66%
- Variants undescribed13%
- Shipping cost revealed late7%
- Hover only variant picker61%
- Add to cart needs JavaScript17%
- Popup blocks the flow10%
- Payment step unreachable9%
| Stall | Stage | Severity | Share of stores |
|---|---|---|---|
| Products have no GTIN or MPN to match againstEvaluate | Evaluate | Medium | |
| Size or colour can only be chosen by hoveringBuy | Buy | High | |
| No sitemap for agents to crawlDiscover | Discover | High | |
| Add to cart needs JavaScript the agent cannot runBuy | Buy | Medium | |
| Variants are not described in structured dataEvaluate | Evaluate | High | |
| A popup blocks the path to the cartBuy | Buy | High | |
| Product pages have no schema.org markupDiscover | Discover | High | |
| Agent could not reach the payment stepBuy | Buy | Critical | |
| Shipping cost cannot be determined before checkoutCompare | Compare | High | |
| Bot protection blocks the agent before the storefront loadsDiscover | Discover | Critical |
What to change first
- No product identifier
Populate the barcode field (GTIN, EAN or UPC) on each variant, which most platforms already have a field for.
- Hover only variant picker
Make each option a real control: a select, a radio input, or a button that responds to a click and to the keyboard.
- No sitemap
Publish /sitemap.xml covering product, collection and policy URLs. Most platforms generate this for you once it is enabled.
4 further stalls were found but withheld: too few stores showed them to report a share without pointing at an individual store.
How it has moved
A trend is drawn once this benchmark has a week of nightly figures under methodology bench-v4. Until then, the figures above are the only ones, and there is no line to read.
Citing this benchmark
These figures are published under CC BY 4.0. Quote them freely, with this attribution:
Ottom, Agentic commerce benchmark across all stores, 27 June 2026 to 25 September 2026, methodology bench-v4. https://ottom.io/benchmarks/all. Licensed CC BY 4.0.
Can AI agents read these stores?
Before an AI agent can compare or buy anything, it has to be let in and find something it can read. These figures come from the first step of every scan, across the 253 online stores Ottom scanned in the 90 days to 25 September 2026, in every category.
- Withheld
Block AI agents in robots.txtEvery store scanned: 253
Withheld. Fewer than 5 of the 253 online stores Ottom scanned in the 90 days to 25 September 2026 showed this, too few to report a share without pointing at a store.
- 6%
Stop agents with bot protectionStores robots.txt lets in: 253
Of the 253 stores Ottom scanned in the 90 days to 25 September 2026 whose robots.txt lets AI agents in, 6% stop an AI agent with bot protection before the storefront loads.
- Withheld
No machine-readable product feedStores an agent could open: 238
Withheld. Fewer than 5 of the 238 stores Ottom scanned in the 90 days to 25 September 2026 whose storefront an AI agent could open showed this, too few to report a share without pointing at a store.
- 3%
No llms.txtStores an agent could open: 238
Of the 238 stores Ottom scanned in the 90 days to 25 September 2026 whose storefront an AI agent could open, 3% publish no llms.txt.
- 10%
No schema.org Product markupStores whose product pages an agent read: 237
Of the 237 stores Ottom scanned in the 90 days to 25 September 2026 whose product pages an AI agent could read, 10% carry no schema.org Product markup on those pages.
- 7%
No shipping costs before checkoutStores whose product pages an agent read: 237
Of the 237 stores Ottom scanned in the 90 days to 25 September 2026 whose product pages an AI agent could read, 7% give no shipping costs an agent can read before checkout.
Each share is counted over the stores that reached the point where it is checked, which the sentence names. A share is withheld when fewer than 5 stores show it. A store enters Ottom’s survey only if its catalogue can be read, so these are mostly stores an agent can reach, and blocking may well be more common across online retail as a whole. The methodology sets out each denominator. The same figures as JSON.
Read next
- How these benchmarks are measured
What counts as a store, what counts as a completed checkout, and why a category with too few stores publishes nothing at all.
- All category benchmarks
Every category Ottom measures, and which of them have enough data to publish.
- Why AI agents abandon carts
The failure modes behind the buy stage, and how they differ from human abandonment.
- AI commerce ROI calculator
Put a revenue figure on the stage where agents stall on your store.