Skip to content
Knowledge & documentationFor every industry

Capture expert knowledge before people leave

Guided interviews record what only one person knows, easy to find by machine or part.

Today

In many companies, key processes depend on a few experienced people. Their knowledge sits in no system, only in their heads, their e-mails and old notes. When they retire, decades of experience leave with them.

An example

At Talwerk Antriebstechnik the master craftsman in gear cutting retires in spring. The agent finds 40 topics on which almost only he answers questions, such as setting up the hobbing machines after a tool change. After 6 interviews the results are linked to 22 machine types and the matching breakdowns in maintenance, every article approved by him.
Fictional company, typical workflow.

Your data, wherever it lives

The knowledge of experienced people sits today in their heads and, scattered, in e-mails and chats, maintenance cases, project folders and handwritten notes. demi reads, understands and links what is written down to machines, parts and processes, with your rights and only from channels you approve. The AI shows which topics depend on a single person and prepares targeted interviews long before anyone leaves; the expert decides what becomes an article.

Where the data lives today

  • E-mails
  • Chats
  • Tickets
  • Folders & drives
  • Scans & paper
Data LayerOne view of everything, with your rights

How demi solves it

Three layers in this order: connect the data first, then the rights, then the AI.

  1. 1 · ConnectData Layer

    The case needs data from e-mail, chat, file share and CMMS. demi replaces nothing and attaches e-mails, chat threads, service and maintenance cases, project folders and interview notes to the machines, parts and processes they concern. demi writes back only after approval.

  2. 2 · PermitPermission Layer

    Rights follow the org chart, sensitive knowledge can be limited to specific teams, and only the channels you release are analyzed. The expert approves every article, and this is recorded.

  3. 3 · ActIntelligence Layer

    The agent spots knowledge risks from who answers which questions and prepares guided interviews. It turns them into articles that quote the expert word for word and point to the original cases.

How it runs

  1. 1

    Connect e-mail, chat, service cases and file share

  2. 2

    Find topics that only one person answers

  3. 3

    Run guided interviews with the experts

  4. 4

    Experts review the articles and approve them

  5. 5

    The knowledge is searchable, linked to machines, parts and cases

What changes

  • Knowledge stays in the company

  • Less dependence on single people

  • Successors get up to speed sooner

What does it bring you?

Check it with your own numbers

Every starting value is an assumption you can change. The calculation sits right below, step by step.

Assumption, editable

Your numbers

Result

+58 hfreed per year

Assumptions and calculation

demi guides the interview and links the knowledge to machines and parts; your experts tell their story.

2 h → 1.4 h per case

How does it run today?

What is already digital saves less: the factor lowers the share demi brings.

How we calculate

8 expert interviews × 2 h × 30% × 1.0 (manual) × 12 months = 58 h

Value per year, before demi costs

Freed per year+58 h

Assumption: labor cost of $48.62 per hour in US manufacturing in June 2026 (BLS), plus workplace, IT and overhead.

= Value per year, before demi costs

+$3,770

gained per year

An estimate, not a promise. The starting values are assumptions for an industrial company with 150 to 500 people.

Combine with more cases and get the package

Questions about this use case

How does demi find out which knowledge is at risk?

The agent looks at who answers on which topics in service cases, tickets and released channels. Topics where the same person is asked almost every time go to the top of the list. The manager sees the list and decides where to start.

Do experts stay in control of their knowledge?

Yes. Every article goes to the interviewed person before it is published. They correct, add or delete and then approve it. Who may read the article later is something you agree on together.

How does an interview work?

The agent prepares questions from real cases, for example a breakdown only the expert could fix. The interview runs in the chat or as a conversation with a colleague who asks the questions. The answers become a draft that the expert reviews.

How is the knowledge found later?

Through the normal search, and directly on the machine or part it concerns. Anyone working on a breakdown sees the matching articles right away. Every article names the expert and the cases it comes from.

What about the works council?

Analyzing who answers which topics touches employee data, so it belongs on the table with the works council early. demi only analyses the channels you release and shows topics, not ratings of people.

See this case with your data

In the demo we show you the workflow on your own systems.

Your configurator

No use cases yet

Your configurator is still empty

Add a use case from any card or case page and see what it brings per year, with the calculation. Free, no sign-up, no commitment: try it out and change any number.

Browse use cases