Answers that are right, because the knowledge behind them is right

AI customer service is only as good as the information it draws on. If your terms are in an old PDF, your opening hours in three different places and your procedures in one colleague's head, your customers will notice. A good knowledge base puts everything in one place, so your AI, your team and your customers all find the same correct answer.

Your parcel arrives tomorrow between 10:00 and 12:00. Prefer a pickup point? You can still switch until 08:00.
from 3 knowledge base cards

The AI puts together an answer from cards in your knowledge base: returns policy, delivery and pickup points.

Sound familiar?

  • Your website says one thing, your team says another, and the chatbot lands somewhere in between.
  • The knowledge mostly lives in the heads of a few experienced colleagues.
  • After a change to your terms or opening hours, you don't know everywhere it needs updating.
  • New starters ask the same questions as your customers.

What it does

  • What your customer experiences

    An answer that is right, whether they ask the chatbot, ask your team or look it up themselves in your help centre.

  • What your team gets back

    One source to fall back on. Less "let me just check", faster onboarding, and an AI that doesn't improvise because the information is simply there.

  • What happens behind the scenes

    Your existing texts, terms, product information and procedures are gathered, cleaned up and set up so the AI can use them well. Where information differs per location, that is recorded separately. Questions the AI can't answer show what is still missing.

What goes into it

  • General information

    Opening hours, contact details, delivery, payment, terms and conditions.

  • Product and service information

    What you offer, who it suits and how the options differ.

  • Procedures

    How a return, cancellation, fault or complaint works, and what the customer needs for it.

  • Per location

    Different opening hours, offering, facilities and contacts.

  • Tone and boundaries

    How you sound, what the AI doesn't talk about and which questions your team always handles itself, outside the AI.

How it works

1Your customer asks"When is the engineer coming?"2The AI looks it upin your knowledge and schedule3Your customer has an answer"Tomorrow between 08:00 and 10:00."Or: your team takes overwith the full conversation
  1. Mapping. Which questions come in and where the answers are now, even if that is in someone's head.

  2. Knowledge and integrations. The information is gathered, cleaned up and recorded. Anything that has to be live, such as an order status, comes from an integration with your system rather than from the knowledge base.

  3. Testing with real questions. With questions your customers actually ask. Where an answer is wrong or missing, the knowledge is filled in.

  4. Live and fine-tuning. Unanswered questions and new topics are reviewed and added regularly. You make changes in one place.

→ In detail: How we work

People stay in control

Example conversation. Customer: I'd like to move my membership to a different location. AI assistant: That's possible. Because your contract is still running, I'll pass you to a colleague. Lisa from the team is taking over Lisa (team): Hi, I'm Lisa. I can see your question, I'll sort it out right away.

  • You are the source

    The AI uses your information and doesn't invent policies that don't exist. If something isn't in the knowledge base, it says so and refers the customer on.

  • You approve

    New or changed knowledge only goes live once you or your team have checked it.

  • Privacy

    The knowledge base holds no customer data. Personal information only comes in through an integration, and only when a question calls for it.

Why XPRT

  • We start with your customer

    We don't just fill in everything. We first set up what your customers really ask.

  • We make it easy to maintain

    One place for each change, so your knowledge base is still right a year from now.

  • You're not locked in

    We choose where your knowledge base works best for your situation: in your helpdesk, on your website or on its own, depending on what you already use.

Frequently asked questions

I already have an FAQ page. Is that enough?

It's a good start. An FAQ page is often too general or doesn't cover every question that comes in. Your existing texts are used and filled in where needed.

How do I keep the knowledge base up to date?

You make changes in one place, and they apply straight away to the AI and your team. Questions the AI can't answer show what is missing.

Can the knowledge differ per location?

Yes. General information sits in one place, and for each location you record what is different, such as opening hours, offering or contact person.

Can my customers search the knowledge base themselves?

Yes, if you want them to. The same knowledge can be a help centre on your website, so customers search for themselves and the AI gives the same answers.

How much work do I need to put in myself?

You supply what you already have and answer questions where something is missing. The gathering, cleaning up and setting up doesn't have to be on your plate.

Want to see how the AI answers from a knowledge base?

Try it at Try the AI live.

Often combined with

→ What the AI handles with that knowledge: What AI handles for you

Where does your knowledge live right now?

Book a demo and bring a few questions your team often gets. You will see how a knowledge base answers them.