Today
Small IT teams in mid-sized companies drown in password questions, printer problems and access requests. Tickets arrive without the device, the error message or the urgency. Real problems wait.
An example
At Talwerk Antriebstechnik a dispatcher writes at 7:40: “VPN not working since this morning.” The agent asks for the device, spots an expired certificate on laptop NB-0417 that is known from ticket 2291 and sends the three-step fix. If that fails, it opens a ticket with the device, the error message and the steps already tried.
Your data, wherever it lives
IT requests arrive today by e-mail, chat and word of mouth, solved cases sit in the ticket system, how-tos in the wiki and in documents on shared drives, devices and accounts in inventory lists and the user directory. demi reads, understands and links them per person, device and error, with your rights. The AI solves standard questions in minutes with a source and turns the rest into complete tickets, and IT decides on access rights and passwords.
Where the data lives today
- E-mails
- Chats
- Tickets
- Wiki & intranet
- Folders & drives
- Office files
- Databases
How demi solves it
Three layers in this order: connect the data first, then the rights, then the AI.
- 1 · ConnectData Layer
Data sources in this case: the ticket system, chat, e-mail and user directory. demi replaces nothing. The case needs data on devices, people, IT documentation and solved tickets. Whatever the agent does not solve itself, demi prepares as a complete ticket; it writes back only after approval.
- 2 · PermitPermission Layer
Users see their own tickets and devices. The agent never changes passwords or access rights itself: access requests go to the responsible approver with a reason, and every step is recorded.
- 3 · ActIntelligence Layer
The agent asks for device, symptom and urgency and answers standard questions from the documentation. Every suggested fix cites the documentation section or the earlier ticket.
How it runs
- 1
Clarify the data sources: ticket system, IT docs, inventory and directory
- 2
Requests come in by chat or e-mail
- 3
The agent asks for the device, the symptom and the urgency
- 4
It solves standard cases with instructions and routes the rest as a complete ticket
- 5
IT sees where instructions are missing and adds them to the docs
What changes
Faster help for everyone
IT works on the real problems
Tickets with every detail
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.
Your numbers
Result
+360 hfreed per year
Assumptions and calculation
demi answers standard questions from the IT docs and opens complete tickets; you solve the rest.
15 min → 10.5 min per case
How we calculate
400 IT requests × 15 min × 30% × 1.0 (manual) × 12 months = 21,600 min = 360 h
Value per year, before demi costs
Freed per year+360 h
Assumption: labour cost of €49.50 per hour in German manufacturing in 2025 (Destatis), plus workplace, IT and overheads.
= Value per year, before demi costs
+€23,400
gained per year
360 h ÷ 1,760 h per role = 0.2 full-time roles
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 packageQuestions about this use case
Can the agent reset passwords or grant access?
Not on its own. It explains the self-service route where one exists and prepares access requests with a reason. The responsible person approves, and every change stays traceable in the ticket.
Which ticket systems are supported?
Your existing ticket system stays, as does a shared IT mailbox. How the data gets from there to demi, we clarify in a conversation. Nothing is replaced. demi writes back only after approval.
Does it learn from solved tickets?
Solved tickets become a source. When the agent finds a similar case, it suggests the earlier fix and links the ticket. IT can turn fixes that come up again and again into documentation articles.
What if our IT documentation has gaps?
The agent only answers where it has a source and opens a ticket otherwise. It shows IT which topics get asked about often without any instructions. That way the documentation grows where it is needed.
