The field — where every AI lab stands in search
Seven AI labs, ranked by estimated organic traffic from live search data. The headline is uncomfortable: Anthropic sits 4th — behind not just OpenAI, but a consumer chatbot and a search company. The leader, OpenAI, pulls 22.5× the organic traffic. This is the gap the rest of the report makes winnable.
Among the AI labs, Anthropic ranks 4th on the metric that compounds: real organic search traffic.
The leaderboard — organic traffic
estimated monthlyCompare Domains — the full side-by-side
The view every competitor analysis starts with: every lab on the same organic metrics, ranked. The one half we don't estimate here is Authority Score, backlinks and referring domains — that's subscription-gated and lives in the upcoming Backlinks module. We never fake an authority number.
Organic metrics · all 7 labs
click a column to sort| Domain | Traffic | Share | Keywords | Top-3 | Top-10 | #1 | Net | Authority | Backlinks |
|---|
Net is keyword positions gained − lost over the tracked window (momentum), not a slice of the current keyword count — that's why a high-volume domain can shed more than it presently ranks for. The signal underneath: every AI lab is net-negative right now. Perplexity has stabilised (−0.5% of its base); the model labs are churning hardest — OpenAI −91.7%, Anthropic −58.2%. The whole category is re-shuffling, which is exactly why the keyword-gap plays below are winnable.
The backlink & domain-authority columns sit behind a separate data subscription — the same paywall Semrush itself sits behind. They're the subject of the upcoming Backlinks module; we mark them honestly rather than estimate a number we can't measure.
Zoom in — Anthropic vs OpenAI
the rival that mattersThe five gaps
■ Anthropic · ■ OpenAIThe multiples
OpenAI doesn't just have more keywords — it converts them to top-3 positions 6× more often. The traffic gap (22.5×) is wider than the keyword gap (3×) precisely because of that ranking quality.
Market position — share & competitive map
This is a consolidated market: of the organic traffic across these seven labs, OpenAI owns 78.5% and Anthropic just 3.5%. The map plots traffic against ranking efficiency — Anthropic sits in the low-traffic corner, below the line on how often its keywords convert to top-3 positions.
Traffic share
organic, estimatedCompetitive position
traffic × ranking efficiencyX = organic traffic · Y = ranking efficiency (share of keywords in the top 3). Leaders rank a lot and rank well (OpenAI); efficient niche ranks well on a small footprint (xAI); Anthropic sits low-left — mid traffic, thin top-3 conversion.
Where your rankings live — the position pyramid
60,150 keywords sounds healthy — until you see where they rank. Anthropic is bottom-heavy: the large majority sit on page 4 and beyond, where they earn almost no traffic. And the trend is negative — a net loss of ~35,000 rankings.
Keywords by position bucket
Anthropic vs OpenAIRanking momentum
Net change = new rankings minus lost. Anthropic is shedding positions faster than it gains them — the inverse of a compounding flywheel.
The keyword gap — terms OpenAI owns and you don't
62 commercial terms where OpenAI ranks in the top 10 and Anthropic is buried or absent. The most damning aren't generic — they're Anthropic's own territory: OpenAI outranks Anthropic on "swe-bench" (your benchmark) and "claude code plugin" (your product name). Sorted by opportunity; click any term for the play.
| Keyword | Volume | CPC | OpenAI | You | Gap | Opportunity |
|---|
AI answer share of voice — who the engines recommend
Beyond Google: when the AI engines themselves answer "which AI lab / model should I use?", who gets named. Here Anthropic is genuinely strong — #2 by share of voice — but OpenAI owns the default: it's named first nearly twice as often.
Share of voice across AI answers
88 replays · 5 enginesThe first-mention gap
Share of voice ≠ default choice. Anthropic is mentioned often (23%), but OpenAI is named first — the recommended option — 48% of the time vs your 27%. Closing that is an entity-authority play.
Google AI Overviews — who gets cited about AI
When Google's AI Overview answers questions about AI, it cites sources — and the single biggest winner isn't a lab at all. Reddit is cited 13×; Anthropic, once. The conversation about you is happening on forums you don't control.
This is the on-page → AI bridge: the same structured-data and authority gaps in the On-Page and Site-Audit modules are why third-party forums, not anthropic.com, own the AI answers about Anthropic.
The displacement playbook — the competitive plays, ranked
Where to attack first. Not "rank for everything" — the highest-leverage moves, ordered by winnability × value.
Competitive intelligence — coverage map
The full competitor-analysis surface, and exactly where this report stands on each — live on real data now, or what the full engagement adds. No silent gaps.
Methodology & provenance
—
Honest gap: the backlink & domain-authority head-to-head is not collected here — it sits behind a separate data subscription and is the subject of the upcoming Backlinks module. We never estimate an authority score we can't measure.