The 2026 State of AI Visibility Index

The 2026 State of AI Visibility Index is the inaugural annual report covering AI citation behavior across the major AI assistants — ChatGPT, Perplexity, Claude, Gemini, and Grok — and the structural patterns that determine which brands get cited at category-defining rates. The Index is built from the same methodology that powers the 10-Point AI Citation Framework and the 40-Point AEO Content Geometry Standard, applied at category-aggregate scale rather than brand-specific scale. The 2026 baseline covers 12 verticals and identifies the citation patterns we'll re-measure quarterly going forward. This article documents the methodology for how the Index is constructed, the initial baseline findings from our first full year of cross-client audits (20+ brands across real estate, local services, B2B SaaS, healthcare, and legal verticals), and the editorial cadence for ongoing publication. Annual updates will be published Q1 each year; quarterly drift reports between annual updates will track per-vertical citation pattern changes.

What does the Index measure?

The Index measures three layers of AI citation behavior:

Layer 1: Per-LLM citation source rosters. For each major AI assistant, the Index documents which sources get cited across category queries — the Pattern A2 directory rosters per vertical, the editorial publications cited consistently, and the brand citation patterns.

Layer 2: Vertical-level signal weight emphasis. For each vertical, the Index documents which dimensions of the 10-Point Framework carry the highest weight. Real estate emphasizes Pattern A2 directories + NAP consistency. Local services emphasizes GBP + emergency-availability signaling. Legal emphasizes Avvo + bar association directories.

Layer 3: Cross-platform citation patterns. The Index documents which signals produce cross-platform citation lift (universal patterns) vs which produce single-platform lift (per-platform tactics).

The Index is a methodology-transparent reference. Every measurement is reproducible by buyers applying the published rubric.

How was the 2026 baseline collected?

The 2026 baseline draws from:

Source 1: 20+ client audit engagements. Across 12 months of cross-client work, we ran the 10-Point Framework audit + per-platform query testing against 20+ brands spanning real estate (luxury + general residential), local services (painting, restoration, roofing, HVAC), B2B SaaS, healthcare, and legal verticals.

Source 2: Published research integration. The SERanking November 2025 study of 300,000 domains — the largest empirical study of AI citation correlation factors to date — provided cross-validation of our internal observations. Position Digital's research on AI Overviews integration provided additional data points.

Source 3: Per-platform query testing. Across each vertical, we ran 30-50 representative buyer queries against ChatGPT, Perplexity, Claude, Gemini, and Grok. Documented which sources got cited per platform per query. Aggregated patterns across verticals.

Source 4: Methodology validation against known outliers. The Westside Luxury sample audit (real, anonymized) and the Tony's Painting CA Inc. methodology validation provided ground-truth scoring against the published rubric.

The combined dataset is small relative to enterprise surveying but methodology-transparent. Every observation in the Index can be traced to a specific source.

What are the headline 2026 baseline findings?

Finding 1: Brand mention frequency is the strongest single predictor. The SERanking 0.334 correlation holds across verticals we tested. Brand mention frequency engineered above competitive averages produces measurable citation lift within 90-180 days.

Finding 2: Pattern A2 directories dominate ChatGPT category queries. Across real estate, local services, legal, and healthcare verticals, ChatGPT cites a small stable directory roster (typically 4-7 sources) per category. Brands absent from the top of their vertical's roster are systematically undercited.

Finding 3: The Two-Track Law holds. ChatGPT and Claude are content-dominant (cite based on content extraction). Perplexity and Gemini are entity-dominant (cite based on entity confirmation via Knowledge Graph, Wikipedia/Wikidata, NAP consistency). Brands optimizing for both tracks see compounding citation lift.

Finding 4: llms.txt has near-zero independent signal. Per cross-vertical testing, presence or absence of llms.txt produces no measurable citation rate difference. We removed it from our scoring framework in March 2026; the rationale is published at /learn/why-we-removed-llms-txt.

Finding 5: Cloudflare WAF defaults block AI bots. A meaningful percentage of audited brands (roughly 30-40%) had Cloudflare default WAF rules blocking GPTBot, ClaudeBot, or Perplexity-User. The Cloudflare GPTBot trap is the most common single technical blocker producing zero citation despite otherwise strong signals.

Finding 6: Schema markup adoption lags. Most audited brands have Organization or LocalBusiness schema but lack Article + Person author schema. Articles with declared author entities get cited at materially higher rates by Claude per platform-specific testing.

Finding 7: Original data publication is rare but durable. Brands publishing quarterly market reports, industry surveys, or proprietary data get cited at sustained rates by Claude specifically. The investment is high (2-4 weeks per quarterly report) but produces a citation moat competitors cannot easily replicate.

What are the per-vertical findings?

The Index breaks findings down by vertical:

Real estate (luxury + general residential): Pattern A2 directory roster is stable (FastExpert, HomeLight, Zillow agent profile, Realtor.com, LuxuryRealEstate.com for luxury). NAP consistency is the most damaging single-dimension gap (split-brain confusion costs Check 6 max at 4/10).

Local services (painting, restoration, roofing, HVAC, plumbing): Pattern A2 directory roster centers on Angi, HomeAdvisor, Houzz, Thumbtack, Yelp + GBP. Restoration vertical adds IICRC + insurance carrier preferred networks.

B2B SaaS: G2, Capterra, TrustRadius dominate Pattern A2 citation. Backlinks weighted higher than for local verticals.

Healthcare: Healthgrades, ZocDoc, Vitals dominate. NPI database verification critical.

Legal: Avvo dominates. State bar association directories carry meaningful weight.

Each vertical sub-section in the Index includes the complete per-vertical Pattern A2 roster + signal weight emphasis + remediation playbook.

What's the editorial cadence going forward?

Annual Index publication. The 2026 State of AI Visibility Index publishes Q4 2026 / Q1 2027. The 2027 Index publishes Q1 2027. Annual updates document year-over-year changes in citation behavior.

Quarterly drift reports. Between annual updates, quarterly drift reports document any per-vertical citation pattern changes — new directories joining preferred rosters, existing directories falling out, per-platform tactic changes.

Methodology version updates. As our methodology evolves (the 10-Point Framework moved from v0.2 to v0.3 in March 2026), changelogs document the rationale + the empirical evidence supporting the change.

The combined cadence keeps the Index relevant without overwhelming buyers with constant change.

How can buyers apply the Index?

Three primary use cases:

Use case 1: Benchmark your brand. Run the 10-Point Framework audit against your own brand. Compare results to the per-vertical baseline in the Index. Identify which dimensions you're above/below typical.

Use case 2: Calibrate methodology investment. Use the per-vertical signal weight emphasis to prioritize remediation investment. If your vertical emphasizes Pattern A2 directories above all else, that's where to start.

Use case 3: Track industry change. Subscribe to the quarterly drift reports to stay current on per-vertical citation pattern changes. Adjust your AEO investment as the patterns shift.

Frequently asked questions

Why publish the Index methodology rather than keeping it proprietary?

The published methodology approach is the core philosophical differentiator of Data for AI Search. Buyers can verify our work; competitors can adopt our patterns; the industry advances faster. The methodology recursion (we publish methodology while applying methodology to publish methodology) is part of the credibility model.

How does the Index handle small-sample bias?

20+ brands across 5 verticals is small for absolute statistical claims. We make directional claims supported by cross-validation against external research (SERanking 300K-domain study, Position Digital research). The Index is the most methodology-transparent annual reference available; it's not the largest empirical study.

What if my vertical isn't covered?

The 2026 baseline covers 12 verticals. Verticals not covered can apply the underlying methodology — the 10-Point Framework applies universally; the per-vertical playbooks are vertical-specific overlays. Buyers in uncovered verticals can run their own audit using the published rubric.

Does the Index recommend specific platforms?

No. The Index is methodology-transparent reference. The best AI Visibility tools in 2026 buyer guide is a separate evaluation framework.

Will the Index be available as a downloadable PDF?

Yes, in Q4 2026 / Q1 2027 when the annual Index publishes. The editorial content on this site is the canonical reference; a downloadable annual PDF will provide an offline reference version.

---

Companion guides: The 10-Point AI Citation Framework · The 40-Point AEO Content Geometry Standard · Pattern A2 directory playbook · Brand mention frequency: the #1 predictor · Why we removed llms.txt from scoring.