AI Knowledge Management · Knowledge Assistant

An assistant that knows your company knowledge.

An AI knowledge assistant answers questions about contracts, processes and products from your own documents, always with source citations. This is what AI knowledge management looks like in everyday use: your team can chat with your documents instead of searching through drives. We start with the knowledge agent pilot, one knowledge area in five days for a fixed price, on SharePoint with Copilot Studio or on your own server.

The problem

The knowledge is there. Just not where it's needed.

It sits in drives, inboxes, and in the heads of two or three people. They get interrupted constantly, and when they're out, things grind to a halt. A knowledge assistant makes this knowledge available to everyone, without anyone having to search for it.

  • 01Looking for answers takes longer than the actual work.
  • 02The same questions always end up with the same colleagues.
  • 03Customers wait for answers that are already documented.
  • 04Onboarding new people depends on word of mouth.
Use cases
01

AI knowledge management: one body of knowledge, many places to use it.

The same knowledge base powers several assistants: internally for your team, in customer contact, and in the inbox. This way, a knowledge assistant covers multiple departments in your company without you having to maintain your knowledge twice.

01

Internal assistant for your team

Your AI knowledge base built from manuals, product data and policies answers your team's questions in seconds, with a link to the source.

Response time Seconds instead of hours
02

FAQ agent for customer inquiries

The agent answers recurring customer questions on the website or by email, from your knowledge base instead of guessing.

Impact Customers get answers faster
03

Inbox with draft replies

Incoming inquiries are recognized and given a ready-made draft reply. Your team reviews and sends it.

Impact The draft is already there
04

Onboarding & induction

New colleagues ask the assistant instead of the grapevine. Knowledge stays in the company, even when people leave.

Impact Faster onboarding
05

Public administration & the public sector

Directives, statutes and procedural knowledge become findable, for staff and, on request, for citizen inquiries.

Impact Team relieved
Technology choice

Your environment decides, not our favorite tool.

We build the knowledge assistant for your company where your data already lives: self-hosted for maximum data sovereignty or directly in the Microsoft tenant.

Self-hosted possible
n8n + open models

Knowledge assistant on your own server. Maximum data sovereignty, freely combinable with your systems and workflows.

Microsoft 365
Copilot on SharePoint

If your knowledge lives in SharePoint and Teams: Copilot agents directly in the tenant, with existing permissions.

Customer contact
Chat & voice

The same knowledge base answers calls and chat inquiries, handing off to your team when things get tricky.

How we work

How the knowledge agent pilot runs.

One knowledge area, five days, a fixed price. After that, you know whether it holds up. Whether we build on SharePoint with Copilot Studio or with n8n on your server, we decide together in the initial consultation, based on your data situation and data protection needs. Example from a project: an item and stock inquiry agent for a medical device supplier, built with Copilot Studio on Dataverse.

1

Day 1: Knowledge area and test questions

Together with your subject-matter expert, we define which knowledge area the assistant may know, such as a manual or your product data. We also write test questions and the answers that would be correct.

2

Days 2 and 3: Build

We connect the assistant to your document store, on SharePoint with Copilot Studio or with n8n on your server. We also define when it answers and when it hands off to a human.

3

Day 4: Testing and refinement

The assistant has to pass the test questions. Wherever it falls short, we refine the knowledge base and set the permissions. The result is recorded in the test protocol.

4

Day 5: Handover

You receive the documentation, and those responsible get a briefing. Whether further knowledge areas or channels are added is your decision afterward.

Pricing
03

One knowledge area at a fixed price, then billed by the day.

You start with one knowledge area and after five days you see whether the assistant holds up. Every further expansion costs 1.000 € net per day.

Expansion

After the pilot

1.000 €net per day
  • Further knowledge areas
  • Further channels, such as website chat or Teams
  • Ongoing support in operation
Book initial consultation

All prices net, plus travel costs and expenses for on-site appointments. No hidden hours. Free initial consultation

More than answers?

When the assistant should also take action.

A knowledge assistant answers questions from your documents. If instead of an answer you need an action, such as an entry in a system or an approval, the agent sprint is right for you: an agent built with Copilot Studio in five days, for a fixed price.

Go to the agent sprint
Free initial consultation

Which questions cost your team hours every week?

In 30 minutes we'll clarify which knowledge area suits the pilot and whether SharePoint or your own server is the better fit. No obligation.

  • Answers with source citations instead of guesswork
  • Your documents stay in your environment
  • Pilot in five days, for a fixed price
Frequently asked questions
05

What managing directors often ask.

What does an AI knowledge assistant cost?

The knowledge agent pilot for one knowledge area over five days costs 5.000 € net. This is a fixed price covering scoping, setup, testing against your test questions, and handover. Any expansion afterward, such as further knowledge areas or channels, is billed at 1.000 € net per day.

For on-site appointments, travel costs and expenses are added. Licenses, for example for Copilot Studio, are not included.

NotebookLM, Copilot, or your own knowledge agent: what fits?

That depends on where your documents live and who is allowed to see them. With NotebookLM you can quickly search individual documents, but the data then sits with Google. Microsoft 365 Copilot works with whatever you're already allowed to see in SharePoint and Teams.

A dedicated knowledge agent pays off when many people need reliable answers with source citations for a fixed knowledge area, in your Microsoft tenant or on your server. You'll find a detailed comparison in our guide to NotebookLM in the enterprise.

Where does the knowledge assistant get its answers from?

Exclusively from the sources you approve: documents, file stores, FAQs, wikis. Every answer names its source, so your team can verify it instead of having to just believe it.

What does the assistant do when it doesn't know something?

It says so. Instead of guessing, it replies that the information isn't in the knowledge base and hands off to a human. This safeguard is a fixed part of every setup.

Do our documents stay with us?

Yes. The assistant runs self-hosted on your server or in your Microsoft tenant. Your content is not used to train third-party models, and a data processing agreement is in place.

How does the knowledge stay up to date?

The knowledge base syncs automatically with your document stores. When a document changes, the assistant knows the new version at the next sync. Outdated sources can be excluded specifically.

How quickly is such an assistant ready to use?

The knowledge agent pilot takes five days for a clearly defined knowledge area, with testing against your test questions and handover. After that, the knowledge base grows step by step.