Why are the wrong metrics so common?
The appeal of the wrong metrics is that they are easy to measure. Counting completed tasks on a board is simple; understanding why a piece of work waited three weeks is not. But easily measured metrics share a problem: they distort behaviour the moment they are measured. Start measuring task count and the team writes smaller tasks, the number rises, and nothing has changed.
| Misleading metric | Behaviour it creates |
|---|---|
| Completed task count | Tasks get written artificially small |
| Hours worked | Looking busy takes priority over finishing work |
| Message / comment count | Unnecessary chatter rises, deep work falls |
| Number of items open at once | Everyone starts many things, nothing finishes |
1. Cycle time
The time between work actually starting and finishing. If you can only choose one metric for productivity, choose this one, because it shows whether the team is genuinely getting faster independently of task size. Use the median rather than the mean: a few long items distort the average badly.
2. Waiting time ratio
How much of a piece of work’s total elapsed time was spent actually being worked on, and how much waiting. The surprising result in most teams is that work spends the majority of its time waiting — for approval, feedback, or another team’s output. The way to speed up a team is usually not to work faster but to remove the reasons for waiting.
3. Throughput
The number of items completed in a given period. On its own it is misleading, but read alongside cycle time it gives a far more reliable basis for capacity planning than estimation. You can answer “how long will this take?” by looking at the last eight weeks of throughput.
4. Work in progress (WIP)
The number of items open at once, directly related to cycle time: as work in progress rises, the time to complete each item lengthens. Counterintuitive as it seems, the relationship is consistent — because a person split across five things cannot make uninterrupted progress on any of them. Tracking WIP shows whether the team is genuinely carrying more than its capacity.
Related pageKanban work tracking software — making per-column pile-ups visible5. Reopened work rate
The proportion of items reopened after being marked done, which measures quality independently of speed. If this rate is rising, the team is not getting faster — it is finishing work incompletely. Tracked alongside cycle time, it immediately shows whether a gain in speed came at the cost of quality.
How should you start measuring?
- Measure for four weeks without setting any target. Measurement that starts with a target corrupts the data from day one.
- Record median values, not averages.
- Share the data with the team. A metric kept secret turns into a trust problem.
- Pick a single improvement and measure for four more weeks. Make three changes at once and you cannot tell which one worked.
The moment you turn a metric into a target, it stops being a good measurement. Use metrics to find direction, not to evaluate performance.
Summary
Measuring team productivity means tracking the flow of work, not the quantity of output. Cycle time, waiting ratio, throughput, work in progress and reopen rate, read together, show where a team is getting stuck and whether it is improving. Record these five for four weeks with no targets; the first opportunity to improve will most likely appear in the waiting times.