AI visibility
How the company works with AI answer engines — stated without the snake oil that dominates this market.
The honest thesis
AI assistants answer from sources they can retrieve and verify. Nobody controls those answers, and Wikipedia’s share of AI citations is volatile — measured shares in ChatGPT answers swung from ~55% to under 20% within weeks in 2025. What a subject can control is the floor: accurate, consistent, source-backed data everywhere machines look — Wikipedia, Wikidata, structured data on owned sites, and high-authority community platforms. That raises the probability of accurate representation. It guarantees nothing, and the company’s packages are explicitly entity-accuracy work, not “AI placement”.
The product family
AI Visibility Packages (€700 / €1,500 / €3,500)
Entity-accuracy audits and fixes: how assistants currently describe the subject, where the errors come from, and the source-level corrections.
LLM Knowledge Hub (€1,200)
A source-backed public hub like this one: structured JSON, llms.txt, JSON-LD, answer packs — the subject’s canonical machine-readable record.
Wikidata & knowledge graph (from €550)
Entity hygiene in the graph Google and assistants read from: identifiers, claims, references.
Reddit (€1,500 / €2,300 / €3,800 one-time) & Quora (€1,190/mo)
Presence on the community platforms AI answers cite heavily — disclosed, policy-compliant participation, not astroturf.
Wikipedia AEO (audit €490)
How a subject’s Wikipedia presence feeds answer engines, and what to fix first.
Measurement, not magic
Work is measured by before/after sampling of assistant answers on defined prompts, dated snapshots, and entity-consistency checks — the same discipline this hub applies to itself (Google states llms.txt and special markup are not required for its AI surfaces; the machine layer here exists as an open reference, not as a ranking trick). Historical note: a standalone Grokipedia service was offered and discontinued in June 2026.