Skip to content
Service & spare partsPumps & ValvesElectronics & SensorsMachinery & Plant EngineeringConveying & Intralogistics+4

Spot frustration before the customer escalates

Repeated calls, cases open for weeks, deadlines set: demi sees the warning signs across every channel and alerts service management.

Today

Escalations seem to come out of nowhere. Yet the customer sent signals for weeks: three calls about the same fault to three different colleagues, a case waiting weeks for a spare part, an e-mail with a deadline, then management in copy. Each colleague sees only his piece, and nobody has the whole picture until the call reaches the managing director.

An example

At Valbrecht the agent flags a university hospital on a Tuesday morning: in five weeks the clinical engineering team called four times about battery faults on syringe pumps, each time reaching a different colleague. A repair has been waiting 26 days for a battery pack, although the contract promises 10 working days, and in her last e-mail the head of clinical engineering sets a deadline for the end of the month, with the commercial director in copy. In passing she mentions that the pumps go out to tender again next year. The agent presents the timeline with all twelve contacts and proposes a call-back by the head of service and five loan units until delivery; the head of service calls before noon.
Fictional company, typical workflow.

Your data, wherever it lives

The signals of an annoyed customer are scattered today: tickets and reopenings in the ticket system, e-mails in several colleagues' inboxes, calls as phone notes, contracts and response times in the CRM, open spare parts in the ERP. demi reads, understands and links them per customer and installed unit, with your rights. AI sees the pattern as it forms, not when the call reaches the managing director; your service management decides what happens next.

Where the data lives today

  • Tickets
  • E-mails
  • Phone & voice
  • CRM
  • ERP
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

    Data sources in this case: ticket system, e-mail, call notes, CRM and ERP. demi replaces nothing and brings together per customer and installed unit what is otherwise spread over colleagues and channels: tickets with their age and reopenings, e-mails and calls, open spare part orders, contracts with agreed response times and running quotes or tenders.

  2. 2 · PermitPermission Layer

    The radar rates customer relationships, never employees: who handled a case feeds into no performance review. Only service mailboxes and tickets you release for it are read. Each warning reaches only people who may see that customer in the CRM.

  3. 3 · ActIntelligence Layer

    The agent checks every customer against the warning signs your service management defines: repeated contacts about the same problem, missed response times, deadlines, threatened penalties or cancellation, management or purchasing in copy, a competitor's name. It also reads the tone, but never raises the alarm on tone alone. From this it builds a timeline per customer with every source and proposes the next step.

How it runs

  1. 1

    demi reads new tickets, e-mails and call notes as they come in and links them to customer and installed unit.

  2. 2

    The agent checks every customer against the warning signs your service management has defined.

  3. 3

    When signals pile up, a timeline forms with every contact, deadline and open point, each with its source.

  4. 4

    Service management gets the warning with a proposed next step, such as a call-back, a loan unit or a visit.

  5. 5

    They decide and act; what is agreed goes into the timeline, and the agent keeps watching the customer.

What changes

  • Trouble caught while a phone call still fixes it

  • The whole picture of a customer in one view, not spread over five colleagues

  • Fewer contract penalties, cancellations and lost tenders

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

+115 hfreed per year

Assumptions and calculation

demi gathers the signals per customer into a sourced timeline; service management decides and acts.

4 h → 2.8 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 escalations × 4 h × 30% × 1.0 (manual) × 12 months = 115 h

Value per year, before demi costs

Freed per year+115 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

+€7,475

gained per year

115 h ÷ 1,760 h per role = 0.1 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 package

Questions about this use case

Which warning signs does demi detect in service?

The ones your service management defines. Typical signs are repeated contacts about the same problem, cases past the agreed response time, reopened tickets, deadlines, threatened penalties or cancellation, management in copy and a competitor's name. Every warning shows which signal triggered it, with the source.

Does the radar monitor my employees?

No. It looks at customers and installed units and rates nobody in your own company. Only service mailboxes and tickets you release are read; you decide which ones beforehand, together with the works council where there is one.

How does the agent read the tone of an e-mail?

Like an experienced colleague: it recognises deadlines, threats, anger and phrases like “as reported three times already”. Tone alone never triggers a warning, only together with signals it can prove, such as deadlines or open cases. Every assessment shows the sentences it rests on.

What happens after a warning in service?

Service management gets the timeline with every source and a proposal, such as a call-back, a loan unit, an express spare part or a site visit. They decide what happens. On request the agent drafts the reply to the customer and keeps watching the case until the signals fade.

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