Service level
What we commit to — in numbers
Availability, response times and credits. Without asterisks, with the measurement method and the exceptions.
- 99.9 % per calendar month
- Response from 1 hour
- Credit without discussion
We commit to an availability of 99.9 % per calendar month — arithmetically up to 45 minutes of downtime a month. To incidents that stop operations we respond within one hour during service hours Mon–Fri 09:00–18:00 (CET); around-the-clock cover is bookable as an add-on. If we miss the committed response, we credit 1/30 of the monthly fee per started hour, up to 30 % a month.
Availability
Measured at the server location, per calendar month.
Announced maintenance does not count as downtime. We schedule the windows at night where the work allows it.
Response times
Response means a human has taken over the incident and is working on it — not an automated acknowledgement.
| Class | What happened | Response in service hours | Around the clock (add-on) |
|---|---|---|---|
| 1 — outage | system down, work is impossible | 1 hour | 30 minutes |
| 2 — restriction | system runs, an essential function is missing | 4 working hours | 2 hours |
| 3 — question, change | adjustment, information, planned work | next working day | next working day |
Service hours: Mon–Fri 09:00–18:00 (CET), excluding public holidays in Brandenburg, Germany. Outside service hours our technical support responds: included in IT management “Full” and in AI operations from the “Act” level (from €129), in other packages for a surcharge.
Credit
If we miss a committed response time, we credit 1/30 of the monthly fee of the affected service per started hour of overrun, up to 30 % of the monthly fee.
The same applies if availability falls below 99.9 % in a calendar month: 1/30 of the monthly fee per started hour below the commitment.
- You request the credit informally within 30 days after the end of the month — an e-mail is enough.
- We offset it against the next invoice.
- Further claims follow the general terms and conditions; the credit is a lump-sum compensation, not a cap on damages caused by intent or gross negligence.
What does not count
- Announced maintenance windows (at least 48 hours in advance, no more than 4 hours a month)
- Force majeure as well as large-scale network or power failures
- Incidents in third-party infrastructure, for example at carriers, where we have no influence
- Causes within the customer’s sphere: own software, own changes, missing cooperation, unpaid services
- Periods in which we are waiting for an approval or information from the customer
Does this fit your operation?
We look at your systems and tell you which level makes sense — and which one you do not need.