Doubao Brand Visibility
Doubao (豆包) is ByteDance's AI assistant and China's largest AI-native app by reported monthly users. Checking a brand's Doubao visibility means measuring, on real consumer clients, whether Doubao mentions, cites, and recommends that brand when people ask category, comparison, and verification questions — and which sources support those answers. API-level checks are not equivalent to verified consumer-surface measurement.
Verified coverage
Real logged-in Doubao mobile-app and web-client measurement is available for snapshots and audits. API telemetry is kept separate and labeled directional.
Product identity
GeoCheckTool is the China AI search measurement product of Super Language (Beijing) Technology Co., Ltd., measuring what China's AI assistants actually say about brands — on real consumer surfaces, with published methods and open data.
Published evidence · bounded scope
Across 16 time-matched pairs of identical questions, Doubao's web and mobile clients displayed different search terms in 15 of 16 pairs and different cited source sets in 14 of 16; read the bounded study and open data. These counts describe that frozen collection and do not prove a permanent surface effect.
Two consumer surfaces, two evidence chains
The phone app and web client must be named separately. A result from one route is not silently generalized to the other, and a bare-model API response is not relabeled as either consumer client.
For every verified run, GeoCheckTool freezes the question, date, account state, surface, and evidence capture so the result can be reviewed later.
The source layer behind Doubao answers
In GeoCheckTool's flagship recommendation-question sample, 119 confirmed marketing-ecosystem citations appeared among 163 total citations, a 73.0%–75.5% sensitivity range. The published report keeps that denominator and sample boundary visible; it does not turn one collection into an industry-wide causal claim.
Verification-style questions can call for different evidence. The practical question is not whether a source category is always good or bad, but whether the answer retrieves facts and sources that support the buyer's actual question.
A reviewable Doubao evidence packet
GeoCheckTool uses dedicated logged-in research accounts, frozen question sets, full-answer capture, source-panel capture, bilingual annotations, and explicit limitations. Every report states the date, batch, account state, and surface before presenting a result.
- China Evidence Snapshot: five frozen brand questions with screenshot evidence and a two-page report.
- Verified audit: twenty frozen questions, per-question evidence packets, citation analysis, and record-layer gap assessment.
- Qwen or Yuanbao can be scoped as a second engine; engine fidelity remains explicit.
Research and measurement paths
The State of AI Search in China
Citation-by-citation flagship report with explicit denominators and open evidence.
OpenBeauty, luxury, and cars
A separate industry note on the entities and sources named in three vertical slices.
OpenVerified China measurement
The $390 Snapshot and the full verified audit on named consumer surfaces.
OpenSurface fidelity methods: Same Question, Same Engine, Two Devices.
Independent research boundary
We publish independent research on Chinese AI search. Paid clients are permanently excluded from our public research samples. Measurement cannot be purchased; our reports state what the engine said, favorable or not.