210 of 500 scanned Shopify storefronts produced access friction before product, policy, or checkout truth could be evaluated.
The finding is an evaluability constraint. It makes no store-specific reach, ranking, recommendation, or revenue claim.
For this report, StoreSteady evaluated a frozen corpus of 500 Shopify storefronts. Freeze: 2026-05-21. Evaluation window: 2026-05-23T16:50:03.820Z to 2026-05-23T23:24:45.338Z.
Run a store-specific scan for crawler access, page collection, and detector evaluability. The result is evidence for your store, not a projection from the benchmark.
See whether your store has agent access friction- Frozen corpus
- 500
- Shopify storefronts entered the public research frame.
- Access friction observed
- 210
- Stores with crawler, robots, or access-friction status before detector truth could be evaluated.
- Crawl failed without classified pages
- 208
- Largest deterministic reason class in the published access-friction table.
- Robots-blocked crawl
- 2
- Stores blocked crawling before enough classified evidence was collected.
Published metric quality
| Metric | Tier | Evaluable N | Finding N | Flagged N | Non-finding N | Rate | Wilson CI | Quality | Caveat |
|---|---|---|---|---|---|---|---|---|---|
Agent access friction observed access_friction_observed | Headline | 500 | 210 | - | - | 42% | 37.8% - 46.4% | deterministic | Deterministic access and crawl-status classification. Access friction is not counted as confirmed breakage. |
Crawl failed without classified pages crawl_failed_without_classified_pages | Supporting | 500 | 208 | - | - | 41.6% | 37.4% - 46.0% | deterministic | Deterministic crawl outcome; not a confirmed product, policy, or checkout finding. |
Robots blocked crawl before classified pages crawl_blocked_by_robots_without_classified_pages | Supporting | 500 | 2 | - | - | 0.4% | 0.11% - 1.45% | deterministic | Robots restriction is reported as access friction, not as breakage. |
Detector non-evaluability not_evaluable_for_detector | Limitation | 500 | 290 | - | - | 58.0% | 53.6% - 62.3% | deterministic | Kept separate from access friction and confirmed findings. |
The useful next step is checking whether your own store creates this same agent-access friction.
Run a store-specific scan for crawler access, page collection, and detector evaluability. The result is evidence for your store, not a projection from the benchmark.
See whether your store has agent access frictionMethodology summary
Selection rule: Frozen external retail Shopify storefront corpus selected before analysis; no merchant-identifying examples are published.
What StoreSteady Measured
StoreSteady measured whether the public scan could collect enough storefront evidence to evaluate product, policy, and checkout truth without bypassing robots rules, login walls, anti-bot controls, or challenge pages.
The headline denominator is the frozen scanned corpus, not the smaller set of stores that were fully evaluable for confirmed Breakage findings.
- Access friction is separate from confirmed breakage.
- Not-evaluable detector outcomes are separate from hard access friction.
- Reason classes are aggregate-only and publish no store names, domains, URLs, crawl text, or raw responses.
What StoreSteady Did Not Measure
This report does not measure recommendation probability, answer placement, placement position, conversion impact, or financial impact.
It also does not claim that any specific AI agent cannot access a named store. The report measures whether StoreSteady could evaluate public evidence under a non-circumventing crawl protocol.
Quality Gates
The access-friction metrics are deterministic status classifications. Judge alpha is not applicable because no LLM judgment is used for these rows.
Every rate uses its own evaluable denominator and Wilson 95% confidence interval. Access friction, detector non-evaluability, runner failures, and confirmed findings remain separate.
Limitations
- Access friction is evaluability friction, not proof of lower ranking, lower recommendation probability, or financial impact.
- The report does not attempt to bypass robots restrictions, anti-bot controls, login walls, checkout gates, or challenge pages.
- Reason classes are aggregate counts only; public artifacts intentionally omit store names, raw domains, raw URLs, raw crawl text, and raw AI responses.
Artifacts
Artifacts are aggregate-only and include no store names, raw domains, raw URLs, raw crawl text, raw AI responses, exact promo codes, private feed/admin evidence, or merchant-identifying examples.
Correction policy
Methodology gaps, reproducible errors, or source corrections can be sent to research@storesteady.com. Corrections are published with the version history in the methodology page and report JSON.