An engine that cannot categorise a business cannot recommend it. This is upstream of every visibility metric: no amount of content moves an answer if the entity record is wrong.
Do AI engines understand who the business is and when it should be considered?
An engine that cannot categorise a business cannot recommend it. This readout shows what each of six AI engines currently believes the business is, and where that belief breaks.
What each engine believes the business is.
- Recognition by engine
- Entity understanding
- Category association
- Confusion detection
- Outdated knowledge
- Source influence
- Recognition decay
What it believes“Midwest industrial fastener and MRO distributor”
Accurate. No divergence from the intended account.
What it believes“An MRO and fastener distributor serving manufacturers”
Accurate. No divergence from the intended account.
What it believes“A regional industrial supplier in the US Midwest”
Describes Ironvale as a hardware retailer rather than an MRO distributor, which removes it from supplier-evaluation answers.
What it believes“Industrial supply company, category association incomplete”
Knows the company exists but not which product categories it carries, so it is omitted from specification-led questions.
What it believes“Unclear, associates the name with unrelated businesses”
Conflates Ironvale Supply with a same-named logistics firm, so category association fails entirely.
What it believes“Not recognized as a distinct business entity”
No stable entity record. Answers reference the category without naming Ironvale at all.
Entity understanding and recommendation presence are separate measures. An engine can resolve the business correctly and still never put it forward, recognition is necessary for a recommendation, not sufficient.
Demonstration environment, seeded organisation Ironvale Supply. Illustrative data, not a customer result.
Signal, evidence, consequence, action.
Every conclusion carries its evidence, its confidence and the intervention it implies, so the readout can be argued with, not just read.
Cindermark Industrial is correctly categorised by all six engines.
Commercial exposureDirectional estimate, not confirmed lost revenue.- Entity category association
Classified as a hardware retailer, not an MRO distributor.
ChatGPT - Entity disambiguation
Conflated with a same-named logistics firm.
Gemini - Entity record
No stable entity record. Category answered without naming any Midwest distributor.
Grok
- Supplier-evaluation coverage 7%
- Source influence 3 independent sources
Publish a single canonical entity description and propagate identical category language to trade directories and structured data.
- Expected movement
- → 55 /100Average recognition score
- Owner and deadline
- Head of Digitalby Sep 30, 2026
Screens this engine provides
- 01AI Recognition Matrix
- 02Entity Understanding
- 03Source Influence Map
- 04Confusion Detector
See what this engine concludes about you.
The demonstration reconstructs one seeded organisation. A live scan runs the same engine against your own business, market and competitors.




