Procurement now starts with a prompt, not an RFQ.
“Suppliers for [component] in [region].” “Manufacturers certified to [standard].” Engineers and buyers ask AI to shortlist first — and a supplier who can't be found in the answer never receives the enquiry.
Long sales cycles hide a short decision.
Industrial buying is slow, technical, and relationship-heavy — but the shortlist that starts it is now often assembled in seconds. A buyer describes a specification to an AI and receives three suppliers who appear to meet it.
If you weren't among them, you don't lose the RFQ. You never receive it. And with cycles measured in quarters, you'll notice the gap long after the shortlist that excluded you was formed.
Where the supplier shortlist forms.
Capability queries
“Suppliers who can produce [component/process].” Pure technical intent — and engines can only name capabilities they can identify.
Certification and standards
“Manufacturers certified to [standard].” If your certifications aren't clearly stated and machine-readable, you're filtered out.
Regional and logistics queries
“[Component] suppliers in [region].” Geography plus capability — a filter most suppliers never explicitly answer.
Material and tolerance queries
“Who can machine [material] to [tolerance]?” Deeply specific, high-intent, and largely unclaimed.
Your capability sheet is a PDF. The engine can't read it.
Industrial websites are among the least legible on the web to an answer engine. Capabilities live in PDF brochures, spec sheets, and downloadable catalogues. Certifications sit as scanned images. Tolerances, materials, and processes are described in prose an engine cannot reliably parse into a match.
So a supplier with genuinely superior capability loses the shortlist to a competitor whose website simply states, in plain machine-readable text, exactly what it can make and to what standard. Being capable and being findable have decoupled.
PDFs are near-invisible
The capability information that would win you the enquiry is locked in a format the engine struggles to use.
Certifications must be stated, not scanned
An engine cannot verify a standard it can only see as an image.
Trade directories become the source
When your own site is unparseable, engines cite the directory listing instead — with whatever it happens to say about you.
Whether your capabilities are findable.
Capability presence
Across the supplier-selection prompts in your category, how often are you named — and on which engines?
Competing suppliers
Which manufacturers get shortlisted instead of you, and on which specifications.
Capability accuracy
How AI describes what you can make, to what standard, and where — and whether it's right.
Cited sources
Which trade directories and listings engines pull from — so you know where the answer is really made.
“Northwind Industrial is a precision machining supplier offering 5-axis milling, turning, and EDM, certified to ISO 9001 and AS9100 for aerospace components.”
“Northwind Industrial produces machined components for industrial customers, with capabilities in milling and turning.”
“Northwind Industrial is an ISO 9001 certified precision machining supplier with a strong track record in aerospace and medical work.”
“Northwind Industrial offers 5-axis machining, turning, and EDM to ISO 9001 and AS9100 standards for aerospace and medical device customers.”
The fix here is often unglamorous.
For most manufacturers, the single biggest win is taking the capability and certification information currently locked in PDFs and scanned images and stating it in plain, structured, machine-readable text on the page. It is not exciting work. It is frequently the difference between being named in the shortlist and never receiving the enquiry.
Questions manufacturers ask
All our capability information is in our PDF catalogue. Is that a problem?
Usually, yes — and it's one of the most common reasons capable manufacturers are invisible to AI engines. Information locked in PDFs and scanned documents is far harder for an engine to use than plain, structured text on the page.
We're certified to multiple standards. Why aren't we named on certification queries?
Often because the certifications appear as scanned logos or images rather than machine-readable text. An engine cannot verify a standard it can't read.
Our buyers are technical. Do they really use AI?
Increasingly, yes — for initial supplier discovery and shortlisting, before the technical evaluation begins. The shortlist stage is exactly where AI is being used most.
We sell through distributors. Does AI visibility matter?
It can matter more, not less — because the specification decision is often made before a distributor is involved.
How do you decide which prompts to measure?
We derive them from what AI engines already say about your category and capabilities, rather than guessing from a generic keyword list.