Monday, before. An operations manager is asked where the team's time goes and gives a confident answer.
She says most of it is on the core process, some on supplier coordination and a bit on queries. That is what it feels like, and it is what anybody in her position would say.
A fortnight of rough tracking shows something different. Answering questions is the largest single category, the core process is smaller than she thought, and the improvement work that has been outstanding since spring received almost no time at all.
Monday before: the estimate is confident and wrong
Not through carelessness. Fragmented time is systematically underestimated, and interruptions feel shorter than they are.
Monday after: the categories are recorded roughly
Six or seven categories, ticked in rough blocks, which takes a minute a day and is the entire method.
Tuesday before: the interruptions are invisible
Each one too small to notice, collectively the largest item, which is the pattern in the enquiries that interrupt operational work.
Tuesday after: the interruptions have a number
Which is usually the moment the conversation about doing something changes from a complaint into a decision.
Wednesday before: the improvement work is assumed to be happening
Slowly, but happening. In reality it receives whatever is left, which is nothing.
Wednesday after: the absence is visible
Zero hours on the thing that would reduce next quarter's workload, which is the most useful single finding of the exercise.
Thursday before: the hiring case is built on a feeling
Everybody is busy, so another person is needed, which is the conclusion in every department in every year.
Thursday after: the hiring case is built on categories
Which frequently shows that the need is not a person but a change to what arrives, and occasionally shows that a person is genuinely needed.
Friday before: nothing can be compared
No baseline, so any change afterwards is argued rather than demonstrated.
Friday after: there is a before
Which is what makes any subsequent improvement measurable, and is the same discipline as proving where closed deals actually came from.
What to categorise
Core work, answering questions, coordination, administration, improvement work and interruptions. Six categories is enough and twelve is too many.
What not to do
Precise time recording, which nobody completes, produces resentment and is not needed to see the shape.
What the measurement itself does
People behave differently while recording, which distorts the numbers slightly and improves the week, and both are acceptable.
What to do with the result
Pick one category and reduce it, rather than attempting a general improvement. A single category addressed properly changes a week, and a broad efficiency drive changes nothing.
What the team should be told
That this is about workload rather than performance, explicitly and beforehand, or the numbers will be distorted by people recording what they think is expected.
Measure for a fortnight before deciding anything
An assistant handling the repetitive questions produces the category that a time audit reveals as the largest, continuously and with a record, which turns the next conversation about capacity into arithmetic rather than a feeling that everybody is busy. The source analysis sits in mapping where repetitive enquiries actually come from, and the status handling in how to use a chatbot for order tracking. See pricing, run the numbers on the chatbot ROI calculator, and the sector view sits under small business.