Maturity survey

Which step you are on, and what your data can answer

Eight questions about what you record, not about how you feel. At the end you will see your step, the list of things we can already compute from your data, and an honest list of what we cannot compute and why. No signup: we ask for an email at the end, and only if you want it.

1 Where is your history of incidents and tickets recorded?

The incident log is where every calculation starts: with no recorded history there is nothing to count.

2 How long is that history without gaps?

The length of the series decides which methods apply: repeats show up over months, periodicity and trend need three months, forecasting and comparison with your own past need a year.

3 Is a priority or severity recorded - and is the field really filled in?

Priority separates an outage from a "replace my mouse" ticket. Without it averages mix things that should never be mixed.

4 Is the affected service or device recorded?

With no object name, availability is counted across the whole estate at once, not for the thing your customer cares about.

5 Are there promised recovery times - even internal ones?

A target is what "on time or not" is measured against. Without one we can say how long repairs took, but not whether that is long.

6 Do you keep a separate log of recurring causes - problems?

A problem log answers what share of your downtime has an owner and a named cause at all.

7 Are changes and releases recorded with a date and time?

A sizeable share of failures follows a change. Without a change log that link can only be guessed at.

8 Are there exportable metrics - load, packet loss, latency, response time?

Metrics allow a check against industry norms and an estimate of when a value will cross its threshold.