About GEO and BASELINE

GEO is the systematic work of getting a brand accurately mentioned, cited and ranked inside the generative answers of AI engines. The core difference from SEO: SEO competes for rank among ten blue links, while GEO competes for the single answer the AI writes directly — and users often decide right there, without clicking through to any website.

SEO optimizes ranking on a results page; AEO (answer engine optimization) overlaps heavily with GEO — both care about the “direct answer.” We use “GEO” to stress that the target is the generated answer inside AI engines (DeepSeek, Doubao, etc.) and the sources it cites. They aren’t mutually exclusive: an AI-friendly website usually does well on SEO too.

Six consumer Chinese AI engines, always: DeepSeek, Doubao, Tongyi, Yuanbao, Kimi and Ernie. Capture goes through real web-interface interaction (driven by Playwright), staying as close as possible to the answer a real user would get — not an idealized response from a public API.

Because a percentage on a small sample is noise. At N=5, “0%” and “20%” differ by a single answer yet read completely differently, and won’t survive scrutiny from a legal team or a competitor. We report only raw counts (X/Y) with a Wilson 95% confidence interval, stating the uncertainty honestly. It’s both integrity and self-protection.

The opposite. Four of the five dimensions are pure code — zero LLM, zero randomness; the only subjective one, “sentiment / frame,” is forced through human review to L1 confidence before entering a report. The same evidence, run by anyone on any day, yields a byte-for-byte identical conclusion. All weight constants are fixed once at design time, literature-anchored, identical for every client, with zero runtime discretion.

Usually about two weeks. You receive: a shareable HTML / PDF white paper (Share of Voice, cited-source distribution, question-by-question breakdown), a site-audit scorecard (0–100 plus agent reachability L0–L3), and a priority-ordered list of recommendations. Every number clicks back to the raw answer screenshot.

Yes. The site audit is an independent step: give us a domain and we produce a three-tier weighted scorecard (basic access / structured data / content and Chinese entities) and point out the first few things to fix. It’s often the first step of an engagement.

We’re very honest about this. Academic benchmarks (C-SEO Bench, NeurIPS 2025; SAGEO Arena) show many tactics are unstable in real conditions — pure prose rewriting can even lose ground (−9%), while structural remodeling is steadier (+17–22%). That’s why we insist on measure first, optimize second, re-measure monthly — rather than selling a one-off “top placement” promise.

The measurement and audit stages are entirely read-only and touch none of your systems. At the “build” stage, every site fix and piece of content is implemented only after you approve it. Our capture data itself is append-only — add-only, never altered — so it stays verifiable.