Inside an AI Visibility Report: What To Actually Check, and Why It Matters

A real report we ran, broken down module by module. plus what it found for one actual business.

Somewhere right now, someone is asking ChatGPT which managed IT provider, which insurance broker, or which physical therapy group to trust.

AI answers instantly, confidently - and it's already decided who gets recommended and who gets skipped. This recommendation can not be made it the chatbot can't actually understand who you are.

That shift is bigger than it sounds. The share of buyers who start their research directly on an AI platform has climbed to 46%, up from just 25% two years ago, while traditional search fell from 43% to 24% over the same stretch. By the time most people reach a website, the decision is often already made: 31% of AI users now say their choice was largely settled before they ever clicked through to a business's own site. ChatGPT alone reaches over 900 million people every week. This isn't a niche channel anymore.

For a growing share of your customers, it's the front door.

So before we ever recommend a fix, we run a full AI Visibility Report (backed by teh nerds over at Bord Labs/ Visibl). Here's exactly what's in it, module by module, and why it's built this way.

Module 1: Visibility Score & Summary

The problem: You can't fix what you haven't measured. Right now, no one at your company can tell you your AI visibility score, because until this report runs, no one has ever calculated it.

That sounds obvious once it's said out loud, but it's the gap almost every business is standing in. Owners know their Google ranking. They know their review count. Almost none of them know whether ChatGPT, Claude, or Gemini would recommend them at all, because that number has never existed for their business before.

This module fixes that. It gives you a real Visibility Score and a Technical AI Readiness score, both benchmarked, both trackable over time. Not a vague "you're doing okay", an actual number you can watch move as the work gets done.

Alongside it, a plain-language executive summary translates the findings into something a founder or a marketing lead can act on in five minutes, not something that requires a technical translator to understand.

Module 2: Technical & Schema Audit

The problem: a website looking fine to a human tells you almost nothing about whether AI can actually read it.

This is the module that explains the gap nobody can see just by looking at the site. Most business owners, when told their AI visibility is weak, assume the fix is "I guess I need a new website."

Usually, that's wrong. AI crawlers — GPTBot, ClaudeBot, PerplexityBot — often have full permission to access a site and still find nothing worth citing, because the structured signals that tell them this is a real business, here's what it does, here's proof it's trustworthy simply aren't present.

The site is reachable. But is it readable in the way that matters?

This module runs the actual technical audit: crawler access and bot permissions, whether llms.txt and agents.md guidance files exist (most sites have neither), and a full schema audit — Organization, Product, Review, AggregateRating, and more — checked present or missing, one by one.

It also looks at content depth: how many indexed pages exist, and whether the site has enough specific, structured content for an engine to cite with confidence.

Module 3: Competitive & Query Gap Analysis

The problem: this is where it stops being abstract. Somewhere, a real customer is typing a real question — "who's the best [service] near me" — and a specific, named competitor is getting the answer instead of you.

Nothing else in the report creates urgency the way this module does, because it's not hypothetical. It shows the actual queries a business is losing, in the real language customers use, and it names who's winning them instead.

Seeing "Competitor B is cited in this exact query, you are not" lands differently than any abstract visibility score ever could.

It turns "we're not very visible" into "we are losing this specific customer, to this specific competitor, right now."

This module benchmarks visibility head-to-head against named competitors, surfaces the specific queries currently being lost, and runs a citation gap analysis — how much of the business's own content is ever cited directly by AI, versus how much of the conversation happens through competitors or third-party sources instead.

Module 4: Roadmap & Citation Tracking

The problem: a diagnosis with no plan just creates overwhelm. Now you know what's wrong, but not what to do first, or whether any of it is even worth doing.

This is the module that decides whether a report actually gets acted on or just sits in an inbox. Every finding gets sequenced by real priority: Critical, High-Upside, Strategic - with a concrete timeline attached, so whoever executes the work (an internal team, or ours) knows exactly where to start and why that's the starting point, not just a list of forty things that all feel equally urgent.

It lays out 7-day, 30-day, and 60–90-day plans that serve as the framework for improving citation traffic.

What This Looked Like For a Real Business

We recently ran this exact report for a local service business with a strong reputation and years of good, recent content. Identity removed here, but the findings are real.

The starting number: 

a 47% Visibility Score and a Technical AI Readiness score of just 31.

Below average, but far from hopeless , the report's own language on this was direct: "a solid foundation with recent content and strong customer reviews, but expertise remains almost completely invisible to AI agents because of critical infrastructure gaps."

The technical finding that mattered most: 

robots.txt allowed every AI crawler in GPTBot, ClaudeBot, PerplexityBot, all welcome, but there was no llms.txt and no agents.md. In plain terms: the door was unlocked, but there was no sign telling anyone what was actually inside worth seeing. Engines could reach the site. They just had no map of what mattered on it.

The schema gap: 

WebSite and FAQPage markup were present — a real start, but Organization, Product, Review, and Aggregate Rating were all missing, meaning the business's own name, services, and star rating existed as plain text a human could read, but not as structured data an AI system could confidently extract and cite.

The competitive reality: 

a named local competitor was already being cited in the majority of the queries this business should have been winning. The business itself appeared in only a fraction of them, because the competitor's site gave AI more to work with.

The plan that followed: 

Enable AI agent discovery in week one: a A fifteen-minute fix with impact within 24–48 hours.

Add Organization, Product, and Review schema in week two. Expand thin service pages with real depth over weeks two and three. Then start tracking citations weekly against the exact queries the business needed to win.

That's the shape of every report we run.

Want to see where your own business stands?  Get your AI Visibility Report — $497  Get the same depth of analysis, built around your actual site, your actual competitors, and the actual questions your customers are asking AI right now.