The AI visibility checklist

When someone asks ChatGPT, Gemini, Perplexity, Copilot or Claude who to hire, the assistant assembles an answer from what it can find, read and trust about you. We have written before about how assistants decide who to recommend. This article turns that into something you can act on: a checklist.

Work through it block by block. None of the items require tricks. Most require honesty, consistency and a bit of housekeeping.

Block 1 — Site clarity

An assistant reads your site like a careful stranger. If a stranger can’t summarise you, neither can the machine.

Your homepage states, in plain words, what you do, who you serve and where you operate — within the first screen, not buried in a slogan.
Every service you offer has its own page with a descriptive title, not a vague umbrella page for everything.
Your About page says who is behind the company, how long you have existed and what you specialise in.
You have removed or rewritten copy that only makes sense to insiders — clever taglines, unexplained jargon, metaphors that hide the actual offer.
Someone who has never met you could explain your business in one sentence after reading your homepage. Test this with a real person.

Block 2 — Consistent facts

Assistants cross-check. When every source states the same facts about you, you become a stable entity worth naming. When sources disagree, you become noise.

Your company name is written identically everywhere: website, Google Business Profile, LinkedIn, directories, invoices’ public face. One spelling, one form.
Your address, phone and service area match across your site and every profile that lists you.
Your description of what you do is consistent — not word-for-word identical, but never contradictory. If your site says “consultancy” and your LinkedIn says “software vendor”, pick one truth.
Old profiles, defunct locations and abandoned pages are updated or deleted. A stale listing is a contradiction with a timestamp.
Rebrands and renames are reflected everywhere, including the places you forgot you were listed.

Block 3 — Third-party authority

A recommendation is an act of trust, and assistants borrow trust from sources they already rely on. This block is earned, not bought — which is exactly why it works.

You are listed, accurately, in the directories and industry registries relevant to your sector and country.
You have genuine reviews on the platforms your customers actually use, and you respond to them.
Reputable publications, associations or partners mention you by name — a guest article, a membership page, a case write-up.
Your work appears in places where real people compare options: community forums, professional groups, comparison pages.
You have never bought fake reviews or planted mentions. Assistants improve at detecting manipulation; the honest route is the only durable one.

Block 4 — Citable content

Assistants prefer to quote pages that answer a question cleanly. Content built for that job earns inclusion; content built to impress rarely does.

You have pages that directly answer the questions your buyers ask — “how much does X cost”, “how long does Y take”, “what is the difference between A and B”.
Each of those pages answers the question near the top, then elaborates — not the reverse.
Your claims are specific and checkable: what you did, for what kind of client, in what context. No vague superlatives.
Headings describe the content beneath them, so a machine can extract the right passage.
You publish at a pace you can sustain with substance, rather than flooding the site with thin filler. One useful page beats ten empty ones.

Block 5 — Technical foundation

None of the above matters if the assistant can’t read you. The good news: most of this overlaps with plain SEO groundwork, so the work pays twice.

Your pages load fast and render their main content as HTML, not only after heavy scripts run.
Your robots.txt does not accidentally block AI crawlers — check it explicitly; many sites block them without knowing.
You use structured data (schema.org) for your organisation, services and articles, so machines get facts in a format they can’t misread.
Every page has an accurate title and meta description, and your site has a clean sitemap.
The site works over HTTPS, without broken links, redirect chains or duplicate versions of the same page.

Block 6 — Measurement

You can’t manage what you never look at. AI visibility can be observed directly, today, for free.

You have asked the major assistants the questions your buyers ask — “who should I hire for X in [your city]?” — and recorded whether you appear.
You checked what the assistants say about you when asked by name, and whether the description is accurate.
You repeat these checks on a regular schedule, because answers shift as models and sources update.
You note which competitors appear in answers where you don’t, and look at what they publish that you don’t.
Findings feed back into the blocks above, instead of sitting in a document nobody reopens. Our guide on how to check whether AI recommends you walks through this step by step.

How to use this list

Don’t try to clear all six blocks in a week. Start with blocks 1 and 2 — clarity and consistency — because they cost nothing but attention and everything else builds on them. Then fix the technical basics, then invest in citable content, and let third-party authority accumulate as a consequence of doing good, visible work.

Progress here is not instant, and anyone promising overnight inclusion in AI answers is selling something else. But every item you tick makes your business easier to find, easier to understand and safer to recommend — for machines and humans alike.

If you want a starting point, run our free self-check. It shows you where you stand before you change anything.