May 19, 2026
5 min read
Metrics anti-patterns
Why Management Should Stop Comparing Velocity Between Teams
Why cross-team velocity comparison creates bad incentives, weak conclusions, and a false sense of management clarity.
Why this comparison keeps coming back
Cross-team velocity comparison is tempting because managers want a compact signal. Velocity is numeric, repeated, and easy to place in a dashboard. That makes it look like an obvious comparison tool.
The problem is that velocity was never designed to be a shared performance currency across teams. It is local by design, which means the neat-looking comparison is usually weaker than it first appears.
Comparison trap
Velocity stops making sense once it is compared across teams that do not share the same context or scale.
Cross-team scorecard
Management comparisons feel objective because they put teams side by side, but the underlying units are not truly standardized.
Different point systems
Relative sizing only works locally, so comparisons quietly assume a level of equivalence the teams never actually had.
Different work shapes
Support load, dependency patterns, and backlog quality can make the same numeric output mean very different things.
Perverse incentives
The scoreboard effect grows quickly once teams realize their local metric is being used for external ranking.
Local metric only
Velocity is healthier when it stays inside the team and gets replaced with better system questions at the management layer.
Why the comparison is structurally weak
Different teams estimate with different local reference points, different work mixes, different backlog quality, and different slicing habits. Even if the numbers look similar on a chart, they are not describing the same underlying unit.
That means the comparison often says more about local point scales than about real delivery capability.
What the comparison starts to break
Once teams know they are being compared, the metric stops being a quiet local planning aid and starts acting like a status score. That changes incentives fast. Point scales drift. Estimation behavior adapts. Velocity becomes less trustworthy precisely when management wants to trust it more.
- Different teams use different local scales.
- Different work mixes make the comparison weaker.
- Comparison pressure encourages gaming behavior.
- The metric starts shaping the system in the wrong direction.
What management should ask instead
The healthier questions are harder, but more useful. Is each team becoming more predictable? Is planning getting more honest? Are risks surfaced earlier? Are tradeoffs clearer? Those questions are closer to real delivery quality than a cross-team velocity ranking.
Managers usually need better context, not a louder scoreboard.
TL;DR
- Velocity is local by design, so cross-team comparison is structurally weak from the start.
- Different scales, work mixes, and slicing habits make the numbers non-comparable in any clean way.
- Once teams know management is comparing them, the incentives around the metric get worse fast.
- Management should ask about predictability, planning honesty, and risk visibility instead.
- Cross-team velocity comparisons create clarity theater because they flatten away the local context that made the metric meaningful in the first place.