If AI can't find you,
nothing else matters.
Discoverable is the first pillar of the P0 Protocol. Before AI can quote your content or recommend your brand, it has to access your site, parse what you do, and place you in the right category. Most brands fail this step without realizing, because it doesn't break the way SEO breaks. There's no ranking drop. There's just absence.
Two halves of the same job.
AI doesn't read your site the way a human does. It needs structured signals it can parse, and it needs enough content to place your brand in a coherent category. Both halves have to be in place for AI to surface your name when someone asks.
The first half is technical access. AI crawlers have to be able to reach your site. Schema has to tell them what kind of business you are. JavaScript can't be hiding the content they need to read. Sitemaps and entity data have to be wired up correctly.
The second half is foundational content coverage. Even if AI can crawl your site, there has to be enough material there to form a picture. Entity definition pages, pillar pages, a real About page, glossary content, and clear comparison or category content. Otherwise AI sees a shell where your brand should be.
Discoverable is the precondition for everything else. Quotable doesn't matter if AI can't reach you. Credible doesn't compound if AI can't connect the citation to a coherent entity.
The four signals of discoverability.
These are the components AI systems check, directly or indirectly, when deciding whether your brand belongs in the recommendation set. Two technical, two content. Each one is independently fixable, and each one is something most brands miss because traditional SEO doesn't surface the gap.
Your robots.txt has to allow the agents AI systems use to read the web. GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and OAI-SearchBot are the baseline. Most brands inherit a default robots.txt that blocks them by accident. The fix is one line per agent. The cost of getting it wrong is being invisible to the systems your buyers are asking.
If your content only exists in JavaScript, most AI crawlers won't see it. They take a snapshot of the rendered HTML, and what isn't there at fetch time isn't read. Server-side rendering, static generation, or selective pre-rendering for AI agents fixes this. The diagnostic is simple. Curl your page, look at the HTML response, and check whether your actual content is in the body or whether it's all empty divs waiting for JS.
Schema is the language AI uses to confirm what your site is about. Organization schema tells it you exist and what kind of business you are. Product schema describes your offerings. Article and FAQ schema let it pull citation-ready answers. BreadcrumbList tells it how your site is structured. Without schema, AI has to guess from context. With schema, it knows.
AI needs enough content on your domain to form a coherent picture. That means a real About page that defines the brand, pillar pages for your core categories, comparison and alternative pages, a glossary that locks definitions, and entity-level pages for the things you do. Thin sites don't get recommended even when the technical access is perfect, because there's nothing for AI to ground a recommendation on.
Discoverable is worth 40 points.
The P0 Score is a 100-point measurement across three categories. AI Discoverability is the largest of the three at 40 points, because if a brand can't be found, the other two categories can't compound. It splits into AI Search Visibility (whether AI systems mention you across the prompts your buyers ask) and AI Competitive Position (how often you appear versus the competitors AI is also returning).
The free AI Visibility Snapshot scores AI Discoverability as one of three top-line dimensions. The full GEO Audit scores it across both sub-dimensions with written findings on each.
Discoverable is the precondition. Without it, work on Quotable and Credible can't compound. The 40-point weighting reflects the structural priority, not just the workload.
The Foundation action.
Foundation is the action that installs Discoverable. It's front-loaded in month one of every Sprint, because everything that follows depends on it being in place. By the end of month one, the technical access is verified, schema is shipped or specified, and the foundational content gaps are filled or scoped.
The work is concrete. We crawl your site the way an AI agent would, identify what's blocked, broken, or invisible, and fix it. We audit and ship schema. We scope and produce the foundational content the brand is missing. The deliverable is a site AI can read.
See how Foundation runs in the Sprint- 01Crawl access verification across the five baseline AI agents.
- 02Schema audit and shipping for Organization, Product, Article, and FAQ.
- 03JS rendering review with SSR or pre-render scope where the gap is material.
- 04Entity definition page if one doesn't exist, plus pillar page scoping.
- 05Glossary build for the brand's category language and definitions.
- 06Sitemap and internal linking review against the new content map.
How brands fail Discoverable.
Most brands don't know they're failing this pillar. SEO traffic looks fine. Rankings hold. But AI doesn't recommend them, and they can't trace why. These are the four patterns we see most often.
Default robots.txt blocking AI agents
The site inherits a robots.txt from a CMS or migration, and it doesn't explicitly allow GPTBot or ClaudeBot. AI systems treat the absence as a soft block. The brand never appears in answers because it was never read.
JS-rendered content invisible to crawlers
The site renders all content in JavaScript. Human visitors see everything. AI agents fetch the HTML, see empty containers, and move on. The brand's depth never gets parsed because it never gets loaded.
No entity definition page
The site has no clear page that says, in plain text, what the brand is and what it does. AI has to guess from the homepage and product pages. Without a confirmed entity, AI can't ground a recommendation.
Thin foundational content
The site has product pages and a blog and nothing else. There's no glossary, no comparison content, no pillar pages. AI has nothing to anchor to when forming a category placement, so it places the brand vaguely or not at all.
What comes after Discoverable.
Discoverable is the first pillar of three. Once AI can find and parse your brand, the next two pillars determine whether AI cites your content directly and whether it recommends you over competitors.
Once AI can find your site, it has to have a reason to quote your content directly in answers. Original research, expert depth, structured answers, and proprietary data published on your own domain. The work that turns your site into the source AI reaches for first.
Read about Quotable → Pillar 03 of 03Authoritative third parties vouch for your brand. News and licensed media, Wikipedia where notability allows, Reddit, expert publications, podcasts, and comparison content. The off-domain work that makes the rest of the web vouch for you.
Read about Credible →Find out where you stand on Discoverable.
The free AI Visibility Snapshot scores AI Discoverability as one of three top-line dimensions. You see whether AI is finding you, where you sit against competitors, and what's keeping you out of the recommendation set. Delivered within three business days.