Current Platform Methodology

How Deeprank Scores Work

Deeprank is about selection and representation. For methodology principles (selection over ranking, declaration over inference), see deeprank.org/methodology.

What the Platform Measures Today

  • AI Readiness Audit: 16 checks total (8 advanced crawl + 8 AI enhanced)
  • Site Check: free 8-check version for quick diagnostics
  • Issue severity and practical recommendations, including fix prompts
  • DeepSeeker: visibility testing across ChatGPT and Perplexity

AI Readiness Audit Structure

8 Advanced Crawl Checks

Readability, headings, meta tags, robots.txt, sitemap, llms.txt, semantic HTML, and accessibility.

8 AI Enhanced Checks

Content quality for AI, information architecture, semantic structure, discovery value, knowledge extraction, context/completeness, uniqueness, and machine interpretability.

What We Deliberately Avoid

Deeprank is not a traditional ranking toolkit. We do not center messaging around legacy ranking playbooks.

We focus on clear declaration, fit/exclusions, machine-readable structure, and measurable AI visibility outcomes.

How DSP Fits the Methodology

The Deeprank Selection Profile (DSP) is the platform's source of truth for identity, audience, fit, exclusions, offerings, and proof. Content tools, snapshots, and snippets are designed to stay aligned with DSP so representation remains consistent over time.

FAQ

Is this a traditional ranking scoring model?

No. Deeprank focuses on AI selection and representation outcomes, not classic ranking-centric workflows.

Where should I start?

Start with Site Check, then run the full AI Readiness Audit, complete your DSP, and use DeepSeeker to measure visibility changes over time.