There's no page two in AI search. You're in the answer or you don't exist.
100 points across 8 signals. You get a number, a tier, and a ranked gap list.
Most gaps are proof and structure — not a new website.
You can hold the #1 spot on Google and still never come up when a buyer asks ChatGPT. They're different systems with different rules, and most businesses are only playing one of them. Here's what that actually looks like:
None of that means your business isn't ready. It means nobody has shown you where you stand — and that's exactly where we begin.
Read the full guide: Why Your Business Doesn't Show Up in AI Search →
The same 8-signal scorecard we use on paid reviews, as a one-page self-scoring checklist. Walk your site, mark each signal, add it up, find your tier. Bring us the score or fix it yourself — either way you'll know where you stand.
100 points, 8 signals, 4 tiers · one-page PDF
100 points across eight categories — weighted toward the proof and clarity AI cares about most. Your total places you on one of four tiers.
The proof AI weighs most heavily.
Can AI tell who you are, what you do, where you work?
Do you cover the territory, or just the pitch?
Can a machine lift a clean, quotable answer?
Do you answer what people actually ask AI?
Can AI reach and read your site at all?
Do you have something only you can say?
Does the rest of the web back you up?
Most websites we review start as Invisible or Findable. The roadmap moves you up.
To get found in ChatGPT, Claude, Perplexity, and Google's AI, your website has to do three things a traditional SEO setup doesn't require: state plainly who you are, what you do, and who you serve so the engine can identify you as an entity; put your actual proof — results, credentials, reviews, methods — on the page rather than in a sales conversation; and structure that content so a machine can extract a clean, quotable answer without guessing. AI engines return one or two recommendations instead of ten links, so being merely present isn't enough. StoryWorks measures all of this on a 100-point AI Readiness Scorecard across 8 signals, then builds a prioritized 30/60/90-day plan to close the gaps.
Each one is scored, each one has a fix, and each one has a deeper guide.
The proof AI weighs most heavily · 20 of 100 points
This is the biggest single block of points, and it's the one almost everyone fails — not because the proof doesn't exist, but because it lives in your head. The jobs you've done, the certifications you hold, the results you've produced, the reason you do it the way you do. A sales call is full of it. Your website has almost none of it. AI can only recommend what it can read, so the work of closing this gap isn't creating anything new — it's publishing what already exists.
Who you are, what you do, where you work · 15 of 100 points
An AI engine has to resolve you as a specific entity — one business, one location, one set of services — before it will put your name in an answer. That means your name, address, and phone consistent everywhere; your services stated in plain language rather than clever ones; your service area named explicitly; and structured data telling machines the same story your homepage tells humans. Ambiguity here quietly caps every other signal, because a machine that isn't sure who you are won't risk recommending you.
Depth across everything you do · 15 of 100 points
Most sites cover the two or three services that make the most money and skip the rest — which means AI sees a fraction of the business. Depth means every service represented, the adjacent questions answered, and enough coverage of your subject that an engine reads you as a source on the topic rather than a brochure about one offer. This is where a hub-and-spoke content structure earns its keep: one page defines the framework, each spoke goes deep on a piece of it.
Structure that makes you quotable · 15 of 100 points
Good content that AI can't extract earns nothing. Extractability means answers that stand on their own without the paragraph above them, headers that state what the section actually answers, definitions in a single self-contained sentence, and no burying the point three paragraphs into a story. It's the least glamorous fifteen points on the scorecard and usually the fastest to win, because the content is already written — it just has to be restructured.
Real buyer questions, answered directly · 10 of 100 points
People ask AI what they'd never type into Google — what does this cost, how long does it take, what do you need from me, what happens if it goes wrong. If your site doesn't answer those, AI finds a competitor's site that does. The fix is a real FAQ built from the questions you actually get on sales calls, each answered directly in the first sentence before any elaboration.
Crawlability, structure, and the files AI looks for · 10 of 100 points
Before any of the content signals matter, the engine has to get in. That means AI crawlers explicitly allowed in robots.txt, an accurate sitemap, JSON-LD schema, content rendered in HTML rather than locked behind scripts, and an llms.txt at your root telling engines who you are and where your important content lives. These are the cheapest points on the board — usually an afternoon of work — and skipping them caps everything else.
Named frameworks, original data, your own process · 10 of 100 points
Interchangeable claims produce interchangeable answers. A named methodology, a scored framework, original numbers from your own work — these give an engine a reason to name you specifically rather than describe your category. It's the reason this page exists in the form it does: the 8 Signals is a named framework with a defined scoring system, and no competitor can be the answer to a question about it.
Reviews, citations, directories, third-party mentions · 5 of 100 points
AI cross-checks. Reviews on Google and third-party sites, consistent citations, directory listings, mentions in places you don't control — this is corroboration, and it's often where an engine goes when your own site is thin. Only five points, but it's the category we most often see scored lowest, and it's the one you can't fix by editing your own website.
We built exactly this structure for a roofing client — a named framework hub, one deep-dive post per promise, all interlinked, each with its own FAQ and schema. Within about an hour of pushing it live, AI engines were naming them first for the best roof warranty in Cincinnati, ahead of two competitors who had been outranking them. A day later the owner reported a new lead from someone who had asked Claude for the best roofing company in Cincinnati.
Our own site's score. We run the scorecard on ourselves, publish the number, and work our own gaps. Ask us what our weakest category is; we'll tell you.
Drop in your website and we'll score it by hand against all eight signals — your score, your tier, your biggest gaps — within two business days. No call required.
Make sure ChatGPT can identify you as a specific business, find your proof on the page, and extract a clean answer. In practice that means consistent entity information, published results and credentials, content structured so a machine can quote it, direct answers to the questions buyers actually ask, and technical access — AI crawlers allowed, schema in place, and an llms.txt at your root. StoryWorks scores all of this on a 100-point scorecard across 8 signals.
The 8 Signals are the eight things AI engines check when deciding which business to recommend, scored by StoryWorks on a 100-point AI Readiness Scorecard: Experience, Expertise, Authority & Trust (20 points), Entity Clarity (15), Topical Depth & Coverage (15), Content Extractability (15), Question & Answer Content (10), Technical AI Accessibility (10), Proprietary Assets & Methodology (10), and Off-Site Presence & Mentions (5). Your total places you on one of four tiers — Invisible, Findable, Citable, or Authoritative.
Traditional SEO gets you into a list of links. AI search optimization gets you into the answer. They overlap — both reward clear, credible, well-structured content — but AI adds entity clarity, on-page proof, machine-extractable answers, and coverage of the questions buyers ask AI. A site can rank first on Google and still never be named by ChatGPT.
GEO is optimizing a website to be quoted and recommended by generative AI tools like ChatGPT, Claude, Perplexity, and Google's AI Overviews. It's the same discipline we call AI search readiness.
Ask it. Open ChatGPT or Perplexity, ask the question your best customer would ask — who's the best at what you do in your city — and see whether your name comes up, and whether what it says about you is accurate. That five-minute test is where every review we run starts.
Faster than SEO. Technical fixes can register within days, and we've seen a full hub-and-spoke content structure change AI answers within hours of publishing. Depth and off-site signals take longer — plan on a 30/60/90-day arc.
Usually not. Most of what the scorecard measures is proof and structure — publishing what you already know, reformatting what's already written, and adding the technical files. A rebuild is the exception, not the starting point.
$2,500, flat rate. It includes your score and tier, a demonstration of what each engine actually says about you, a category-by-category breakdown, a one-hour gap interview, your AI Archive, and a prioritized 30/60/90-day roadmap. Implementation is quoted separately. The initial score is free.
Yes. Download the scorecard, walk your own site, and start with the technical points and your FAQ — those are the fastest wins. The part that's hard to do alone is the proof: getting what's in your head onto the page is what the gap interview exists for.

Clarity-first marketing for growth-minded local businesses. StoryBrand messaging, local SEO, and websites built to get found.

Certified StoryBrand Guide.
EXPLORE
FREE RESOURCES
© 2026 StoryWorks Website Design & Marketing. All rights reserved.