The Diagnostic: How to Know If Your Business Is Invisible

I started writing these essays because of an electrician.

He had a 5.0 rating on Google. Genuinely excellent work. Customers left detailed reviews describing exactly the kind of job he did well. By every measure that used to matter (a clean profile, a working website, a track record in the community) he was doing everything right.

But when I asked ChatGPT who to call for electrical work in his city, his name did not come up. When I checked Google’s AI Overview for the same query, it was not there either. Competitors with weaker ratings and less impressive reviews were being recommended. He was not.

The problem was not his work. The problem was that AI search could not confirm he existed in the way it needed to.

That was the first essay. The rest of this book has been an attempt to explain why that happens, what it means, and what to do about it.

What the book has been arguing

Eight essays in, the argument is the same one I had in my head after that electrician audit, just fleshed out from eight different angles.

AI search engines do not evaluate businesses the way human visitors do. A customer reads your copy and decides whether to trust you. An AI engine reads your site for structured signals and decides whether it can safely reference you. Those are different jobs, and most small business websites are built only for the first one.

The visibility gap this creates is structural, not competitive. It is not that your competitors are better at marketing. It is that the rules changed and the brief nobody gave you, the one that said your site needs a machine-readable identity layer, was never part of the design conversation.

The essays have covered the specific pieces: schema markup and entity signals (Ch.3 and Ch.6), review volume as a ranking input distinct from rating (Ch.4), FAQ schema as an AI citation mechanism even after rich results disappeared (Ch.5), why running ads before building the foundation costs more than it earns (Ch.7), how specialization survives the commoditization of the generalist layer (Ch.8).

This final chapter is the checklist. The six questions that tell you whether your business has the foundation in place.

The six-question diagnostic

These are the checks I run at the start of every engagement. You can run most of them yourself in under thirty minutes.

1. Ask ChatGPT who to hire in your category and city.

Open ChatGPT and type: “Who are the best [your trade] in [your city]?” Then type your business name and ask what it knows about you.

Two different results. The first tells you whether you have category AEO, whether AI engines include you when someone asks for your service in your area. The second tells you whether you have branded AEO, whether AI engines know who you are when asked directly.

Most businesses have some branded AEO and almost no category AEO. The goal is both.

2. Check your Google Business Profile for completeness.

Open your profile and look at four things: Is your service area defined with specific cities and radius, not just your address? Are your service categories accurate and complete? Do you have at least a paragraph of business description with the specific services you offer named? Is your phone number, address, and website consistent with what appears on your website?

Inconsistency between your profile and your site is a trust signal problem. AI engines cross-reference. Discrepancies reduce confidence.

3. Check your review count against the top three Map Pack competitors.

Search your category in Google Maps. Look at the review counts, not just the ratings, for the three businesses that appear in the Map Pack. Write down the average. That is your target, not any industry benchmark.

BrightLocal’s research confirms review signals account for 16% of local SEO ranking factors. A perfect rating with six reviews does not carry the same weight as 4.8 stars with 180 reviews. Volume is a separate signal from sentiment.

4. Check your robots.txt file for AI crawler access.

Go to yourdomain.com/robots.txt in a browser. Look for any line that says Disallow: / under a User-agent block. If GPTBot, ClaudeBot, or PerplexityBot appear under a Disallow rule, or if there is a blanket User-agent: * with Disallow: /, those crawlers cannot index your site.

A blocked robots.txt is the fastest path to AI invisibility I see on audits. The business can rank on Google, have great reviews, and still be completely absent from AI-generated answers because the crawlers that feed those systems were locked out during a site build and nobody noticed.

5. View your site’s page source and search for “schema”.

Right-click any page on your site, select View Page Source, and use Ctrl+F to search for “schema” or “application/ld+json.” If nothing appears, your site has no structured data.

No schema means no machine-readable identity. Your site is text to an AI engine, not an entity. The minimum for a local business is a LocalBusiness block with your name, address, phone number, service area, and service categories. Without it, AI engines have to infer who you are from your copy, and inference is not confidence.

6. Check whether your site has service-specific pages.

Count how many pages on your site are dedicated to individual services, not a services overview page, but individual pages for individual things you do. An electrician’s site should have a page for panel upgrades, a page for EV charger installation, a page for residential rewiring. A page that lists all services in bullet points is not the same thing.

Service-specific pages are how topical authority gets built. They are also the pages that match specific queries in AI retrieval. An overview page answers the question “what does this business do.” A service page answers the question “what does a panel upgrade cost and how long does it take in Vancouver.” That is the question the retrieval filter is looking for.

What to do with the results

If you ran these six checks and found problems on all six, that is normal for a small business website built before 2024. The brief that was followed when the site was built did not include any of this.

Start with the robots.txt check. It is the only one that is a hard gate. If AI crawlers are blocked, nothing else matters until that is fixed.

Then schema. LocalBusiness structured data can be added without rebuilding the site. It does not change how the site looks to humans. It changes what AI engines can read.

Then service-specific content. This is the slowest part, but also the part that compounds. Every page you add is another surface for retrieval. Competitors who have been at this for years have thirty to a hundred of these pages. Starting now is earlier than starting later.

Reviews and profile completeness run in parallel with everything else. They do not require a developer. They require a system: a follow-up message after every completed job with a direct link to your review form, sent consistently.

The version of invisible that matters

There is a version of invisible that is obvious: no website, no Google Business Profile, a business that does not exist online in any form. Most small businesses reading this are not that version.

The version that actually costs customers in 2026 is subtler. The site exists. The profile exists. Google can find the business. But when someone asks an AI assistant for a recommendation, which is how a growing share of buying decisions now start, the business is not in the answer. It does not appear in the candidate pool. It is structurally absent from a conversation it does not even know is happening.

That is the version this book has been about.

The good news is that it is fixable. The signals AI engines rely on are buildable. They are not the province of large brands or large budgets. A local electrician with six reviews and no schema can close that gap. The question is whether he knows the gap exists.

That is what I set out to explain.

---

Ian Chang is the founder of SyncSpark, a Vancouver agency specializing in SEO, AEO, and AI search optimization. Run the free Website Scorecard to get a structured assessment of where your site stands across six signal categories.