Knowledge & data
Knowledge catalog
Roughly what does it cost? Is it right for us too? How does it work? Your business answers these questions every day – but none of the answers are written down anywhere you could look them up.
You'll get …
- 1
One organised knowledge base
Services, price ranges, common questions, processes and references sit in one place with a clean structure – instead of scattered across website copy, PDFs and people's heads.
- 2
The same answer everywhere
Website, companion, sales and AI search engines draw on the same checked answers – no more three different prices in three places.
- 3
Gaps and outdated content come to light
Building it shows what's missing, contradictory or no longer true – and with clear responsibilities, the catalog stays up to date.
01Intro
The knowledge catalog collects the answers to your customers' questions: services, prices or price ranges, processes, references, special cases – checked, dated, with someone responsible, in one single place. From there they're used the same way everywhere: on the website, in the companion, in quotes, on the phone and for AI search engines. Whoever asks, wherever they ask, the answer is the same. And building it shows you along the way what's missing, contradictory or out of date. How you work internally – standards, projects, experience – doesn't belong here but in the AI knowledge system.
Worth knowing: this is about putting things in order, not technical wizardry – structure first, then the tools. And a catalog only stays alive if someone maintains it: the project sets out who is responsible and how often things are updated, otherwise it goes stale.
02How it works
- 01
Gather the knowledge
Website, PDFs, price lists, email replies, client questions: what's where? Even knowledge that's only ever passed on by word of mouth gets put on the table.
- 02
Define the structure
What kinds of knowledge do you have – services, FAQs, prices, processes, references? Each gets clear fields: "it's in the text somewhere" becomes "it's in field X, and Y keeps it up to date".
- 03
Fill it and tidy up as you go
Putting everything in its place brings it all to light: the FAQ answer that contradicts the quote PDF, the price from two years ago, the service with no description.
- 04
Connect the channels
Website and companion, quote templates and price lists, sales on the phone, ads or a display in the shop, structured data for search and AI: every channel reads from the same source, wherever it's needed.
- 05
Make upkeep a routine
Who updates what, how often, with which review date? Keep it small and realistic – one area that stays accurate beats ten that go stale.
03Stages
The catalog grows one topic at a time – nobody organises their whole company in one go.
Starter: one topic
One topic is captured properly, usually the one people ask about most – services and prices, for example, or the website content. It feeds the first channel; more follow later.
Next step: all the important topics
Every topic customers ask about is in the catalog: services, prices, processes, references, special cases. Website, companion, sales and quotes all read from it and give the same answer.
Full build: a living database
The catalog keeps growing, with a maintenance process, owners and a date for each building block. New channels plug in without anyone copying text: ads, price lists, the display in the shop window, AI search engines.
04Try it yourself
The catalogue of the fictional “Lichtwerk Signage” firm. Right now every place answers differently. Connect them to the catalogue, then change a value in the block and save – website, companion, quote and phone all update at once.
Every place keeps its own version of the answer.
This is how the catalogue is built: building blocks with fixed fields – each with an update date and an owner – and output places that draw from them. No place keeps its own copy: when a value changes, everything that uses it follows. AI search engines get the same facts as structured data.
Pick a building block in the “try it out” tab – it lights up here along with the places that use it.
05Technical implementation
- Storage to suit you: a Markdown structure, CMS collections, Airtable/Notion or a database – what matters is clear content types with fields.
- Typical kinds of knowledge: services, FAQs, prices and price ranges, references, target groups, processes, decision criteria.
- It can be used in many ways: website pages, structured data (Schema.org), context for the companion, quote templates, exports for ads or displays, an internal lookup view.
- AI helps with sorting, spotting gaps and wording – you check the facts.
- Vector search (RAG) only makes sense at large scale – a clean structure with direct context is usually enough.
- Every area of knowledge has an owner and a review date – upkeep is built into the model, not an afterthought.
for context · This shows how something like this usually gets built – not how it has to be built for you. What actually fits is worked out in the project: around your existing setup, your team, and what stays easy to maintain.