GEO basics
SEO optimises a page's rank in a list of links, aiming for clicks. GEO (Generative Engine Optimization) optimises how discoverable, understandable, trustworthy and citable a company is inside AI-generated answers — the goal is to be in the answer itself. They share technical foundations (crawlability, structured data, content quality), but GEO additionally requires governed entity facts, a machine-readable surface (llms.txt, Markdown mirrors) and a third-party mention ecosystem. A site that ranks well is not automatically visible to AI: engines cite the companies whose facts they can fetch, parse and verify. The concrete differences sit in four dimensions — goal, object, content structure and evaluation — compared item by item in the body below.
Read the full answer →Going global
For Chinese companies going global, the visibility gap is rarely content volume — it is the machine-readable surface. Our audit of 30 Chinese outbound companies found machine-readability the weakest category (69.2/100 average): 17 had no JSON-LD on the homepage, 15 had a missing or thin sitemap, 13 had a missing or duplicated H1. The path has four steps: audit bot reachability and machine-readability first; govern entity facts so Chinese and English statements agree; publish question-shaped content built on real overseas buyer questions; then earn third-party mentions (media, communities, partners) and verify with periodic resampling against a fixed question set.
Read the full answer →AI search
Five changes. (1) Bot reachability: robots.txt explicitly allows AI crawlers and live fetches are not stopped by firewalls or challenge pages. (2) A machine-readable surface: an llms.txt fact file (optional aid), a complete sitemap, Markdown versions of core pages (optional). (3) Entity structure: JSON-LD naming entity types such as Organization, clean title/H1 discipline, hreflang pairing on bilingual sites. (4) Question-shaped content: pages organised around buyer questions, short answers first, FAQ schema attached. (5) Measurability: analytics that separate human and bot traffic and attribute AI-referred inquiries. Negency's own site was rebuilt to exactly this standard, and the public audit tool scores against it.
Read the full answer →AI search
llms.txt is a plain-text brief at the site root written for AI systems: who the company is, what it does, where the core pages live, and the canonical facts and definitions. It does not by itself raise citation rates — in Ahrefs' May 2026 study of 137,210 domains, 97% of published llms.txt files received zero requests that month, so its direct effect is near zero — but it is the cheapest fact-governance move available: one file anchors a single version of the truth that crawlers can rely on. Verdict: worth adding, useless alone; it works as part of entity-fact governance, structured data and question-shaped content. Negency's llms.txt is a full fact bible: company definition, offering boundaries, sourced cases and citation rules.
Read the full answer →Enterprise Agent
Four preparations, and we advise against starting without any of them. (1) Scenario selection: pilot on a workflow that is frequent, comparatively rule-clear and verifiable — not the most glamorous one. (2) Knowledge and data foundations: govern the knowledge base first (deduplication, versioning, permissions); a dirty knowledge base only scales errors. (3) Governance design: permission boundaries, human approval points and audit logs decided before launch, not after. (4) A measurement baseline: record the manual process's time and error rates before the pilot, or value can never be proven afterwards. Negency's method is a small controlled pilot, verified before scaling, with governance as a design input rather than a patch.
Read the full answer →Enterprise Agent
Because an agent differs from ordinary software in one decisive way: it chooses its own next actions, and its errors amplify at machine speed. Governance answers four questions — which data it may touch (permissions), which actions require a human (approval points), what it did (audit trail), and how it stops (kill switch). Approval points are not distrust of the technology; they keep accountability on the human side: external sends, money, deletions and commitments pass a human, everything else flows automatically. The EU AI Act likewise requires human oversight for higher-risk systems — governing early is the cheapest compliance path.
Read the full answer →GEO basics
Four metrics. (1) Mention rate: fix a set of questions your buyers actually ask (about 30, bilingual), sample the major engines weekly, and count in how many answers the brand appears. (2) Citation mix: of the sources cited, how many are your own properties versus third parties. (3) Sentiment and accuracy: whether mentions recommend, neutrally describe or criticise you — and whether the facts are right. (4) AI-referred conversion: visits and inquiries arriving via AI channels, attributed separately. One caution: single-day data is noise — citation rates fluctuate up to 35% within a day, so run a full quarter of baseline before reading trends. Negency's stats backend tables bot crawling and AI-referred leads separately precisely to ground this measurement.
Read the full answer →AI search
Separate the crawlers by purpose before deciding: search-index crawlers (OAI-SearchBot, PerplexityBot) and user-triggered fetches (ChatGPT-User) decide whether you can appear in AI answers at all; training crawlers (GPTBot and peers) are a separate licensing choice — a company may allow search while declining training. For a website whose job is winning customers, explicitly allow at least the search and user agents: if they cannot fetch your pages, AI answers cannot cite you. Whether to allow training follows your own content-licensing policy. In our audit of 30 Chinese outbound companies, only 3 blocked all three major AI crawlers — and those three have capped their own visibility ceiling. Allowing is not exposure: keep admin paths closed, rate-limit by real IP against abuse, and declare training/search/input permissions separately via Content-Signal. Negency's own robots policy — allow everything, close only admin paths, rate-limit for protection — is public and reusable.
Read the full answer →Choosing a vendor
A traditional SEO agency optimises rank in a list of links and reports rankings and traffic curves. Negency builds a GEO system — bot reachability, entity facts, machine-readable surface, question-shaped content, third-party mentions and inquiry attribution as one chain — and reports AI mention rates and attributable inquiries. Three concrete differences: we sell no link packages; we promise no rankings (AI answers have no fixed positions to promise); and every figure is traceable (cases carry sources, our own numbers come from a queryable stats backend). If your buyers only click search links, traditional SEO suffices. If they have started asking AI, you need the other system.
Read the full answer →GEO basics
Two horizons: technical fixes (crawler access, machine-readable surface, consistent entity facts) show up in audit scores and crawl logs within 2–4 weeks; actually being mentioned and cited in AI answers typically takes 2–3 months or more, because it depends on accumulating third-party mentions and on engine re-crawl cycles. Any promise of being recommended by ChatGPT within two weeks is not credible — mention rates can only be verified by periodic sampling against a fixed question set.
Read the full answer →Choosing a vendor
There is no single market price; the structure has three layers. Self-serve monitoring tools run from a few hundred to a few thousand dollars per month. A one-off diagnostic is fixed-price (Negency's human-reviewed diagnostic package is RMB 19,800, including a manual review of audit data and a 90-day roadmap). Ongoing GEO engagements are quoted by scope — number of markets, languages, content volume and depth of third-party mention work. Diagnose first, scope second: a total price quoted without data is a guess dressed as a quote.
Read the full answer →AI search
An audit checks roughly 30 signals along one chain — can AI find you, read you and trust you — in five groups: crawler reachability (are bots allowed in, is the sitemap complete), machine-readable surface (llms.txt, Markdown versions, fetch success), entity facts (is company information structured and consistent everywhere), content structure (question-shaped organisation, heading hierarchy, short answers first) and citation readiness (do cases and numbers have public, checkable sources). Negency's free audit scores each group and points at specific pages and problems, not just a single total.
Read the full answer →Going global
Your website needs a complete version in every language your target market uses to ask AI — engines match citations to the language of the question, so a Chinese-only site is structurally disadvantaged in English answers. Three management rules keep it sane: one page tree, one record, language fields maintained in pairs (two separate trees always drift apart); hreflang annotations so engines know which version serves which market; and a fact pass over every translation, because numbers, dates and credentials must match item for item.
Read the full answer →Going global
Two separate tracks. For the website, it is mostly GDPR: collecting emails, form and behavioural data needs a lawful basis, the privacy policy must be real (not a copied template), consent must be evidenced, and data-subject rights (access, erasure) need a working process. For AI applications, the EU AI Act grades by risk: most enterprise applications are limited-risk, where the core duties are transparency (users must know they are dealing with AI) and record-keeping; high-risk uses such as recruitment, credit or critical infrastructure carry far heavier duties and must be classified at project start. The Data Act and NIS2 also affect connected products and digital supply chains. This page is a practical framework, not legal advice — formal conclusions need European counsel.
Read the full answer →AI search
Structured data is the 'fact sheet' machines read: it tells crawlers unambiguously which company an entity is, what a page is (an FAQ, a service, a report), and what the facts inside are. Its importance is asymmetrical — it does not buy you citations, but its absence causes a class of real failures: AI reads your facts wrong, merges you with a namesake company, or simply cannot parse your pages. In Negency's sample of 30 Chinese companies going global, 17 lacked JSON-LD entirely — the most common major gap in the whole machine-readable group.
Read the full answer →Enterprise Agent
The best first use cases share three traits: high repetition, clear rules and reviewable output. The safest starts are knowledge retrieval with drafting, structured repetitive work (lead triage, report generation, meeting notes) and monitoring with pre-processing — agents fetch, organise and warn; humans keep the decisions. Do not start with autonomous outward promises (auto-quoting, auto-paying), irreversible operations or processes whose rules nobody can articulate. Two or three scenes run well for a month is worth more than ten scenes demoed once.
Read the full answer →AI search
Do not start with complaints to the AI companies — the wrong fact almost always comes from a wrong or conflicting source the engine is reading. The fix order: find which page the wrong version lives on (often your own outdated page, a platform listing or Wikipedia); make one canonical, machine-readable version of the truth easy to find (llms.txt plus JSON-LD on the official site); correct the third-party sources in citation order; then verify by re-sampling the same questions over the following weeks. Vendor feedback channels exist but are slow — treat them as escalation, not the fix.
Read the full answer →Choosing a vendor
It depends on who performs better against one yardstick. Overseas agencies are stronger on local media and native content, and weaker on source verification on the China side, cross-language fact control and communication speed with headquarters. A China-based team is the mirror image: strong on facts and communication, and it stands or falls on whether it truly operates in the target market's language and media ecosystem. Negency's positioning is deliberately hybrid — a Chinese team, entity facts verified at source, English content actually published on LinkedIn and our own channels, and results measured with one fixed sampling method.
Read the full answer →Choosing a vendor
Different unit economics. Large consultancies sell senior time and process: broad scope, long cycles, deliverables that lean to reports, and a minimum engagement size mid-market companies rarely fit. A specialist team like Negency sells one narrow capability delivered as running systems and published pages (the audit tool on this site is itself the product demo), with fixed-price diagnostics and regular measurement loops. The choice is not who is smarter; it is whether the job is enterprise-wide transformation (buy the scale player) or making one capability real, fast and provable (buy the specialist).
Read the full answer →GEO basics
GEO (Generative Engine Optimization) is the systematic practice of making a company discoverable, correctly understood, trusted and cited inside AI-generated answers from ChatGPT, Perplexity, Gemini and their peers. Its goal is not a higher rank in a list of links, but a place inside the answer itself: when a customer asks an AI, are you there, described correctly, and cited from your own pages?
Read the full answer →Going global
When a company expanding overseas is invisible in AI search, the cause is rarely one thing: it is five stacked problems — cannot get in, cannot be read, cannot be recognised, does not match the question, and nobody mentions them. This 17-item self-check, benchmarked against Negency's measurement of 30 global-facing corporate websites, locates which layer you are stuck on and what to fix first.
Read the full answer →Choosing a vendor
GEO has no official leaderboard and its results lag by weeks, which leaves vendors a lot of room for talk. These 12 due-diligence questions — grouped into measurement, method, evidence and contract risk — each come with what a credible answer sounds like and the red flags to walk away from, so you can pin a GEO provider down before you sign.
Read the full answer →Choosing a vendor
What exactly do you get for a ¥19,800 diagnostic? This page opens up the deliverable of Negency's AI search visibility diagnostic package — the scorecard, the itemised findings list and the 90-day roadmap, at delivery granularity. Sample figures come from the published median profile of our 30-company going-global benchmark, not from any real client; a real engagement measures your own domain and three competitors, item by item.
Read the full answer →Choosing a vendor
GEO monitoring tools — Profound, Otterly and their peers — measure how AI engines mention, cite and describe your brand. They change none of it. A GEO agency changes the inputs: your machine-readable website surface, your entity facts, your citable content and your third-party mentions. Measurement tells you where you stand; remediation moves you. Confusing the two is the most common way GEO budgets get wasted.
Read the full answer →Choosing a vendor
Almost every GEO provider fits one of four kinds: a global SEO or content agency with a GEO line, a GEO-native boutique, a China-based going-global specialist, or a large consultancy. Their slogans sound alike; what actually differs is five testable things — how they measure AI answers, which engines and languages they cover, whether they can repair the technical machine-readable layer, whether they can earn third-party mentions, and how transparent the engagement terms are. Pick the kind by your bottleneck first, then diligence the individual firm.
Read the full answer →Going global
Buyers ask AI before visiting your site. The four AI-visibility losses consumer electronics brands suffer most, and the three fixes to make first.
Read the full answer →Going global
Buyers ask AI before they talk procurement. The four point losses medical device brands suffer most, and the three fixes to make first: registration facts in page text, locked model naming, citable clinical evidence.
Read the full answer →Going global
Buyers ask AI before they talk procurement. The four point losses industrial equipment brands suffer most, and the three fixes to make first: selection specs in page text, locked model naming, service and case pages.
Read the full answer →