The Question to Ask First

Most arguments about time theft are arguments about a total nobody has produced. One question surfaces that.

The question

Which of the six, how much, and how do you know?

Ask it of any proposal, any report and any figure.

What the answers mean

"Buddy punching, four instances we identified, from the incident records." A real, measured problem. Proportionate response available.

"Break overruns, a median of three minutes over, from the break data." Measured, and mostly structural. The response is arrangements.

"All of it, we think it's about five percent." An industry figure. Not about your organisation, and the conversation should stop until it is measured.

"People on the remote team might not be working." Not measurable without disproportionate surveillance, and the concern is output visibility.

"System anomalies, one hundred and forty last month." Not time theft at all. A fault list.

Why it works

It separates the measured from the assumed without anyone having to say which they are.

It produces a number, and the number is frequently smaller than the response proposed.

And it identifies which of the six you actually have, which determines whether the response is a purchase, a policy or a repair.

The second question

And what does it cost in the other direction?

Run from the same records, over the same month.

If the answer has never been produced, that is itself the finding, and producing it changes the conversation more than any argument about fairness.

Using both

The first decides whether there is a problem and which one.

The second decides what the net position actually is.

Neither requires buying anything, and both are harder to answer than any question a vendor will ask you.

The collection in one line

Measure both directions, fix the arrangements that cause most of it, handle the genuine cases properly, and stop calling the rest theft.

Everything else here is the detail of doing that.

Apply it to reports too

Not only to proposals.

A report headed "time theft" should be readable as: which of the six, how much, how measured.

If it cannot be read that way, it is a total with categories folded into it, including ones that are system faults and ones that were estimated.

Ask for the breakdown. It usually shrinks considerably, and what remains is actionable in a way the total was not.

A practical configuration prompt

During configuration, use how the product works to prompt questions about identifiers, ownership and output. Treat the page as a starting point and document each assumption.