The symptom
A small task keeps coming back: fixing an entry, hunting for information, explaining a step. Each occurrence looks reasonable, but you wonder whether an improvement would help.
The signal
Note a repetition that matters, then examine its frequency, its consequences, and the possible reasons before you choose a response.
What’s going on
A repetition makes a question visible. It can signal a defect upstream, a need for training, or ordinary activity for the service. Explaining several times to different people can be necessary; a document does not always replace being there.
At a community group’s desk, several people ask which document to attach to a file. The information may be missing from the form. It may also be hard to understand, out of reach in the format offered, or different depending on the case. Counting the questions helps you choose where to look, without establishing the cause on its own.
Frequency is not the only criterion. A rare mistake with heavy consequences can call for immediate action. A frequent, short task can stay cheaper to do than to automate. Do not wait for a third occurrence if a protection or some help is already needed; conversely, three occurrences do not prove a new system is useful.
Keep a proportionate record: a few cases, their context, the rough time, and the effects observed. Say when an estimate is crude. Someone starting out can note a case with a peer to learn what to look at. An experienced person can examine a settled habit, without assuming the people following it have stopped thinking.
Check this
Choose a single repetition over a period that suits your work. Note what recurs and what varies. If you are watching other people’s work, explain the purpose and agree how much time to spend.
At the end, compare two options: carry on as is, or try a bounded improvement. Add the cost of preparing, passing on, and maintaining that improvement.
Have the people concerned decide on the change. At the next comparable use, check whether the effort or the difficulty went down without moving the problem onto someone else. Carrying on by hand can still be the right choice.
From where you sit
- Support: tell a recurring question apart from a cause already verified.
- Design: watch what varies before proposing a shared component.
- Operations: compare frequency, severity, and the cost of an improvement.
- Management: set aside a little time for observation without imposing a permanent inventory.
To discuss
Which repetition deserves examining, and which repeated activity is still delivering something necessary?