PATTERN

AI inquiry on your website

Website content → vector database → MCP tools → swappable LLM → streaming sidebar.

AI search grounded in your own content — no hallucinated answers, just statements with source attribution.

When this pattern fits

Three pain points where the effort pays off.

  1. Callers ask the same question over and over — and the answers have long been somewhere on the website, just not findable.

  2. The website has grown over years; search returns hits, but no answers in complete sentences.

  3. Generic chatbots on the website hallucinate or simply do not know the in-house terminology.

How the pattern is built

Five steps, from input to filing.

  1. Website content

    Crawl plus structured extraction.

  2. Vector DB

    Semantic indexing per language.

  3. MCP tools

    Search, booking, escalation.

  4. Language model

    Swappable per task.

  5. Streaming sidebar

    Word-by-word answer, with sources.

Live example

From real operations.

Customer Shorin-Ryu Seibukan Karate-Union Deutschland e.V. WEBSITE-AUSKUNFT

A long-grown association website with hundreds of pages, several dojo locations, a seminar calendar, and a federation structure becomes a friendly Karate-Cat in the sidebar — that doesn't invent answers, but pulls each one from the actual website content.
See the case

What is deliberately out of scope

Three places where we leave it to a person.

  • Binding legal or medical advice. The AI points; the responsible person decides.

  • Answers without citations. Every statement leads back to the website content that supports it.

  • Lead generation at any cost. Anyone who really needs a person gets a person.

Typical project scope

Discovery → pilot → rollout.

  1. Discovery

    One to two weeks. We understand your concrete bottleneck, look at the available sources and decide together whether the pattern is really the right answer here.

  2. Pilot

    Four to eight weeks. A scoped variant on real data, with a small user group. Dry-run before every write, clearly defined success criteria.

  3. Rollout

    From two months onwards, fluid. Extension to further sources, teams or sites. Hand-over to operations with documented architecture and training for those responsible.

Talk to us

Two doors, one address.

Specific bottleneck?

Let us talk for 30 minutes about your use case.

No obligation, no cost, with concrete next steps at the end.

Book a 30-minute call

Your own AI platform?

See CompanyWizard live in action.

Demo with your own data is possible. We bring the pseudonymisation set up and ready.

Request a demo