# The Industrial AI Visibility Benchmark 2026

> An anonymised measurement of what AI assistants can actually retrieve from 984 manufacturer, distributor and industrial-commerce catalogues.

**Published:** 2026-08-11 · **Fieldwork:** 2026-08-10 · **Edition:** 2026.1  
**Canonical:** https://partsgraph.ai/research/ai-visibility-benchmark-2026  
**Data:** https://partsgraph.ai/research/ai-visibility-benchmark-2026/data.csv (CC BY 4.0)

## The finding

The industrial web is not ready to be read by machines, and the gap is not confined to small suppliers.

Across 984 catalogues the median score was 50 out of 100. 671 sites — 68% — scored below 55, the point at which an assistant is reconstructing a product rather than reading it. None reached the top band; the best result recorded anywhere in the cohort was 80.

## Headline figures

- **984 catalogues** — manufacturer, distributor and industrial-commerce sites measured on what AI assistants can actually retrieve from them.
- **50 median score** — Half the market scores at or below 50 out of 100. The mean and the median sit within a point of each other: this is not a few laggards dragging an otherwise healthy field down.
- **0 sites graded A** — Not one site in 984 cleared the top band. The best result recorded was 80.
- **68% mostly opaque or worse** — 671 of 984 sites scored below 55, the point at which an assistant is reconstructing the product rather than reading it.

## Distribution

Scores run 28–80. Median 50. The middle half of the market falls between 34 and 57; the 95th percentile reaches only 69.

| Grade | Score | Meaning | Sites | Share |
| --- | --- | --- | ---: | ---: |
| A | 85–100 | Agent-ready | 0 | 0% |
| B | 70–84 | Largely readable | 39 | 4% |
| C | 55–69 | Partially readable | 274 | 28% |
| D | 40–54 | Mostly opaque | 356 | 36% |
| F | 0–39 | Effectively invisible | 315 | 32% |

## What was measured

Each catalogue was scored on five dimensions, stated here as the questions they ask. Weightings and probe design are not published — see Limitations.

1. **Crawler policy** — Are the assistants' own crawlers permitted, and is that permission stated deliberately rather than inherited from a default?
2. **Reachability** — When a crawler asks for a product page, does it get one — without a challenge page, a redirect loop, or a timeout?
3. **Readability** — Is the product's substance present in the served HTML, or assembled in the browser after load where a non-rendering crawler will never see it?
4. **Catalogue coverage** — Can the catalogue be enumerated and traversed, and do the product pages carry structured product data rather than prose alone?
5. **Agent surfaces** — Beyond the human website, is anything published for machines — a stated AI policy, a plain-text mirror, a declared tool endpoint?

## Machine-facing surfaces

- **State an explicit AI-crawler policy:** 193 of 984 (20%). Naming the assistants' crawlers, rather than leaving them to a wildcard rule written years before those crawlers existed.
- **Publish an llms.txt:** 90 of 984 (9%). The proposed plain-text index for language models.
- **Serve a plain-text or Markdown mirror:** 35 of 984 (4%). A machine-readable rendering of a page that is otherwise delivered as an application.
- **Advertise a tool endpoint for agents:** 0 of 984 (0%). Zero. Not a rounding artefact — no site in the cohort published a discoverable agent tool interface.

## Reachability

389 sites — 40% — served no product page that could be read at all. Where a catalogue could be enumerated, the cohort exposed 894,308 product URLs directly, against an estimated ~54,983,275 catalogue URLs once sitemap indexes are scaled out.

## Method

Fieldwork on 2026-08-10 against live public sites, from multiple network vantage points, using only routes a member of the public could take. Nothing was logged in, no authentication was attempted, and no rate limit was deliberately exceeded. The cohort combines hand-curated majors across electronics distribution, electrical wholesale, MRO, building products and component manufacturing with a broader filtered sample of industrial commerce, deduplicated by company.

The scoring formula, dimension weightings, probe design and sampling frame are not published; they are the commercial method. This is a real limit on independent replication and is stated rather than dressed up.

## Limitations

- **A blocked probe is a floor, not a reading.** 285 of 984 results (29%) are low-confidence because the site challenged, throttled or refused the request. Those scores are the lowest the site could be scoring, not a measurement of it.
- **Absence and silence are different facts.** A 404 is evidence a thing is not there; a timeout, empty 202 or challenge page is evidence of nothing. The two are never collapsed into one "not found".
- **One day, one set of network positions.** A snapshot, not a longitudinal study.
- **It measures retrievability, not commercial outcome.** A high score means an assistant can read the catalogue — not that it recommends the part, that the data is correct, or that the buyer converts.
- **The sourced half carries a false-positive rate.** A hand-checked sample found a minority of domains that were not genuinely industrial commerce. Headline figures are reported over the whole cohort without excluding them, which makes the result marginally conservative.

## How to cite

```
Partsgraph (2026). The Industrial AI Visibility Benchmark 2026: an anonymised measurement of 984 manufacturer and distributor catalogues. Edition 2026.1, fieldwork 2026-08-10. https://partsgraph.ai/research/ai-visibility-benchmark-2026
```

## References

1. [Vercel — “The Rise of the AI Crawler”. Analysis of ~1.3 billion AI-crawler fetches over one month, including 569M by GPTBot and 370M by Claude. Published 17 December 2024.](https://vercel.com/blog/the-rise-of-the-ai-crawler)
2. [Forrester (B. Winters) — “How GenAI And Trust Are Reshaping B2B Buying In 2026”, Forbes, 2 February 2026.](https://www.forbes.com/sites/forrester/2026/02/02/how-genai-and-trust-are-reshaping-b2b-buying-in-2026/)
3. [Forrester — “The State Of Business Buying, 2026”, drawing on a Buyers’ Journey Survey of close to 18,000 business buyers. Press release, 21 January 2026.](https://www.forrester.com/press-newsroom/forrester-2026-the-state-of-business-buying/)
4. [IETF RFC 9309 — Robots Exclusion Protocol. The 2022 standardisation of robots.txt.](https://www.rfc-editor.org/rfc/rfc9309.html)
5. [The /llms.txt proposal.](https://llmstxt.org/)
6. [Model Context Protocol — specification and documentation.](https://modelcontextprotocol.io/)
7. [schema.org — the Product type.](https://schema.org/Product)
8. [Regulation (EU) 2024/1781 — Ecodesign for Sustainable Products Regulation, the framework introducing the Digital Product Passport. In force 18 July 2024.](https://eur-lex.europa.eu/eli/reg/2024/1781/oj/eng)
9. [ETIM International — the classification model for technical products.](https://www.etim-international.com/)

---

Partsgraph — the agent-ready parts data layer. Free AI-visibility grader: https://partsgraph.ai/audit