How is 3LA different from classic SEO?

Classic SEO measures rankings on a SERP. 3LA measures whether ChatGPT, Perplexity, Claude and Gemini cite, name and recommend you when users ask intent-driven questions — even on queries that never appear in Google Search Console.

Which AI models does 3LA test against?

Layer 3 audits run live citation probes against ChatGPT, Perplexity, Claude and Gemini, with additional Ahrefs Brand Radar signals to detect mentions across the AI-indexed web.

What is a "Dark Opportunity"?

A Dark Opportunity is a high-intent query — typically buying, comparison or recommendation phrasing — where an AI model names a competitor and you are absent. These queries usually do not appear in Google Search Console because the user never visits Google.

How long does a full 3-layer audit take?

A free Layer 1 + Layer 2 audit completes in roughly 60 seconds for a single URL. Layer 3 brand probes typically run in 2–5 minutes depending on category complexity and number of competitors.

Can I use 3LA in Norwegian and other languages?

Yes. The 3LA platform and audits are bilingual (Norwegian + English) and Layer 2 explicitly measures hreflang coverage and multilingual reach. Layer 3 probes can be run in the language of your target audience.

Do I need to be a paying customer to see my 3-layer score?

No. The Layer 1 + Layer 2 audit is free for any URL. Layer 3 (Brand & Dark Opportunities), Ahrefs LinkProfile data and weekly Reclaim reports are part of the Pro, Team and Enterprise plans.

The 3-Layer Approach Framework·v2026.1·2026-02-01·Nettpilot AS · 3LA ResearchLanguage:ENNO

The 3-Layer Approach to AI Search Optimization.

A canonical framework for measuring whether AI models can read, cite and recommend your brand. Three layers, one score, designed for the generative search era.

Run free audit Download whitepaper

Framework overview

Three layers. One score. Built for generative search.

Each layer builds on the one below. Without Layer 1, content has no audience. Without Layer 2, the brand has no source. Without Layer 3, the brand has no mirror.

Technical AI Readiness

Can crawlers and LLMs actually read your site?

Layer 01

Content Authority

Is your content citable, atomic and source-worthy?

Layer 02

Brand & Dark Opportunities

When models mention your category — who do they name?

Layer 03

01

Layer 01

Technical AI Readiness

Layer 1 is the foundation. If GPTBot, PerplexityBot, ClaudeBot and Googlebot cannot fetch and parse your pages cleanly, every downstream investment in content and brand is wasted. We measure this objectively with crawler logs, robots.txt, llms.txt, semantic HTML and Core Web Vitals.

What it measures: How readable, fast and structurally clean your site is to AI crawlers and search engines.
Why it matters: Without crawl access and parseable HTML, your content is invisible to ChatGPT, Perplexity, Claude and Google AI Overviews — no matter how good it is.
How 3LA scores it: We audit robots.txt, llms.txt, sitemap, semantic markup, schema coverage, render-blocking JS and Core Web Vitals — then map each finding to a fixable action.

Metrics we measure

▸robots.txt + llms.txt directives (AI crawler allow/deny posture)
▸Schema coverage (Article, FAQPage, BreadcrumbList, Organization)
▸Semantic HTML density (<article>, <section>, <dl>, <h2-h3>)
▸Core Web Vitals (LCP, INP, CLS)
▸Render-blocking JavaScript and client-side hydration penalties

Audit Layer 1

02

Layer 02

Content Authority

Once a model can read your page, the next question is whether it would ever cite you. Layer 2 measures atomicity, factual density, third-party signals and the kind of structural cues (definitions, lists, tables, sources) that make LLMs select your text as a quote.

What it measures: How atomic, well-sourced and structurally citable each page is for generative AI.
Why it matters: LLMs extract self-contained facts, not whole pages. If your content is not atomic and cited by others, you will not appear in answers.
How 3LA scores it: We measure atomic content units, definition density, external citation signals (Ahrefs LinkProfile), heading hierarchy and the presence of FAQ/HowTo/Definition markup.

Metrics we measure

▸Atomic Content Units (ACUs) — self-contained, citable claims per page
▸External citation authority via Ahrefs LinkProfile (refdomains, rating)
▸Definition density (<dl>/<dt>/<dd>, glossary blocks)
▸Hreflang + multilingual reach for non-English queries
▸FAQ/HowTo/Definition schema presence

Audit Layer 2

03

Layer 03

Brand & Dark Opportunities

Layer 3 is the part traditional SEO tools cannot see. We probe ChatGPT, Perplexity, Claude and Gemini with real category prompts, capture which brands they cite, and surface the "dark opportunities" — high-intent queries where you are absent and a competitor is named.

What it measures: Brand visibility, share-of-voice and "dark opportunities" inside live generative answers across 4+ AI models.
Why it matters: A query you cannot see in Google Search Console may already be sending intent to your competitor through ChatGPT. Layer 3 is your only mirror into that.
How 3LA scores it: We run live citation probes per category prompt, compute Citation Rate vs. competitors, cross-reference Ahrefs Brand Radar, and flag queries where you are missing from the answer.

Metrics we measure

▸Citation Rate across ChatGPT, Perplexity, Claude, Gemini
▸Share-of-voice vs. named competitors per prompt
▸Dark Opportunities — high-intent prompts where you are absent
▸Ahrefs Brand Radar signals (mentions across the AI-indexed web)
▸Sentiment + accuracy of how models describe your brand

Audit Layer 3

FAQ

Frequently asked questions.

3LA is an AI search optimization framework that scores any URL across three layers: Technical AI Readiness (can AI read you), Content Authority (would AI cite you) and Brand & Dark Opportunities (does AI name you when it answers your category). It is the operational successor to traditional SEO for the generative search era.