Click2.ai®Free audit

Industry report · CLX-GEO-2026-06

AI search ranking factors,
2026 edition.

The 9 ranking signals AI search engines actually use, scored across ChatGPT, Perplexity, Claude, Gemini, Google AI Overviews, and Microsoft Copilot. Per-engine weighting, consensus signals, and the surface mechanics that change every few months.

AI search ranking factors 2026 industry report

Executive summary

9 signals total. 5 are consensus signals that work everywhere. 4 are engine-specific.

Build for the 5 consensus signals first — complete schema, FAQPage structure, quotable sentences, named authors, third-party mentions. This is the highest-leverage GEO work because it benefits every engine simultaneously. Then layer engine-specific weighting on top.

The 9 signals

In order of cross-engine importance.

Signal 1 · Consensus

Crawler access in robots.txt

Engine-specific bots: GPTBot · ClaudeBot · PerplexityBot · Google-Extended · Applebot-Extended. Block any = invisible to that engine.

Signal 2 · Consensus

Complete JSON-LD schema

Organization · LocalBusiness · Service · FAQPage · Article. All present, all valid. AI engines treat schema-complete pages as higher-confidence sources.

Signal 3 · Consensus

Quotable phrasing

Short factual declarative sentences with numbers. "Migration in 2–4 weeks at 50–70% lower cost" beats "passionate about modern architectures."

Signal 4 · Consensus

FAQPage Q&A blocks

Real questions, direct answers, FAQPage JSON-LD wrapping. Highest-extraction format across all engines.

Signal 5 · Consensus

Named authors + dates

Visible datePublished, dateModified, Person schema for authors. Provenance-free content loses citations.

Signal 6 · Engine-specific

Third-party brand mentions

Highest weight for ChatGPT and Claude. Medium for Perplexity. Lower for AI Overviews (Google substitutes its own authority signals).

Signal 7 · Engine-specific

Entity consistency (NAP)

Highest weight for AI Overviews and ChatGPT (local recommendations). Medium for Claude and Perplexity.

Signal 8 · Engine-specific

Topical density

Pillar + supporting + city × vertical matrix. Highest weight for AI Overviews. Medium for ChatGPT and Perplexity.

Signal 9 · Engine-specific

Freshness (datePublished/Modified)

Highest weight for Perplexity (fastest-reflecting engine). Medium for ChatGPT Search. Lower for Claude default mode.

Per-engine top 3

What each engine weights most.

ChatGPT

Crawler access · brand mentions · quotable phrasing

FAQ extraction is high. llms.txt is parsed. Search mode uses Bing partnership.

Perplexity

Crawler access · freshness · source diversity

Highest citation density of any engine (5–10 sources per answer). Fastest-reflecting.

Claude

Named authors + dates · internal consistency · crawler access

Trust-signal-dominant. Conservatively declines unverifiable citations.

Gemini

Google ranking · schema completeness · Google-Extended access

Inherits most Google ranking signals. Schema completeness is a separate strong factor.

AI Overviews

Google ranking · schema completeness · E-E-A-T

Page must generally rank well for the underlying query first. Then schema + author/publisher trust signals.

Copilot

Bing ranking · schema completeness · brand entity recognition

Uses Bing index. Schema-complete + branded entity beats unbranded thin content.

FAQ · 8 questions

Methodology and limits.

What are the 9 signals?

Crawler access · complete schema · quotable phrasing · FAQPage structure · named authors + dates · third-party mentions · entity consistency · topical density · freshness.

What matters most for ChatGPT?

Crawler access · brand mentions · quotable phrasing.

What matters most for Perplexity?

Crawler access · freshness · source diversity. Highest citation density of any engine.

What matters most for Claude?

Named authors + dates · internal consistency · crawler access. Trust-signal-dominant.

What matters most for AI Overviews?

Underlying Google ranking · schema · E-E-A-T. Page generally must rank well first.

Are there signals that work everywhere?

5 consensus signals: schema · FAQPage · quotable sentences · named authors + dates · third-party mentions. Highest-leverage GEO work.

Is llms.txt a ranking signal?

Observed to be parsed by ChatGPT, Perplexity, Claude as entity-grounding. Not a formal ranking signal documented by any engine. Helps most on small-to-medium sites.

How fast do these signals change?

Surface (crawler names, robots.txt, model versions) — every few months. Underlying 9 signals — stable since first RAG generation, likely stable through 2027.

Read next

Engine playbooks.

Primary sources & references

Where the numbers come from.

This report cites only first-party publisher disclosures, official developer documentation, and crawler-confirmed behavior from the Click2.ai measurement panel. No second-hand statistics.

Google Developers

Google's common crawlers →

Authoritative list of Google's crawler user agents and robots.txt tokens, including Googlebot, Google-Extended (Gemini training + grounding), GoogleOther, Google-CloudVertexBot, Googlebot-News.

IndexNow protocol

IndexNow.org →

Cross-engine URL submission protocol used by Bing, Yandex, Naver, Seznam. Underpins the freshness signal observed for ChatGPT Search and Bing Copilot.

Click2.ai · case study

83 pages, schema-complete, indexed in 24 hours →

First-party measurement data from the Click2.ai program: page count, schema coverage, IndexNow HTTP 200 confirmation, 20-prompt monthly measurement panel. Dataset schema, CC-BY 4.0.

Apply the report

Free SEO + GEO report.

Your site scored on all 9 signals across all 6 engines. 5 business days. No cost.