Scrum and Agile teams are often encouraged to measure their work.
Velocity, throughput, cycle time, burndown charts and delivery forecasts can all offer useful signals.
They can show patterns, reveal bottlenecks and help teams ask better questions.
But most Agile work happens in complex environments, where outcomes emerge, context shifts and people respond to the system around them.
In that kind of setting, metrics do not simply describe reality. They also shape behaviour.
That is where trouble starts.
Metrics are not neutral
In simple work, a metric can often track something clear and repeatable.
In complex work, things are different.
Cause and effect are harder to isolate. Outcomes emerge over time. Context matters. Human judgement matters.
When you choose a metric in that kind of environment, you are not measuring a fixed machine. You are measuring people who will respond to being measured.
That response can change the system itself.
What gets measured starts to drive behaviour
Once a metric becomes visible, people naturally start paying attention to it.
If it becomes a target, it starts influencing decisions.
That can lead to unhelpful patterns such as:
- choosing work that improves the number rather than the outcome
- avoiding valuable work that is hard to measure
- splitting work unnaturally to make progress look faster
- reporting activity instead of learning
- protecting the metric rather than telling the truth
The metric may still look healthy, even while reality becomes less healthy.
Complex work always contains things that numbers miss
Some of the most important parts of Agile work are hard to reduce to a dashboard.
For example:
- trust within the team
- quality of collaboration
- clarity of purpose
- willingness to raise problems early
- learning from experiments
- value delivered to users
These things matter deeply, but they do not always show up neatly in charts.
If you focus too narrowly on what can be counted, you can lose sight of what counts.
Use metrics as signals, not verdicts
Metrics are often most useful when they start conversations.
They can help a team ask:
- What might this be telling us?
- What might it be hiding?
- What has changed in the system around this number?
- Does this match what we are seeing in real work?
That is a healthier use of metrics in complex environments.
The goal is not to find one number that tells the truth.
The goal is to stay in contact with reality.
Meaning comes before measurement
Good teams do not begin with “What can we measure?”
They begin with questions such as:
- What are we trying to learn?
- What outcome are we trying to improve?
- What would meaningful progress look like?
- What evidence would help us understand that better?
When meaning comes first, metrics can support judgement.
When metrics come first, judgement often gets pushed aside.
A better way to think about metrics
In complex systems, metrics should support learning, not control.
Metrics should help teams notice patterns, explore assumptions and make wiser decisions.
Used well, metrics can illuminate.
Used badly, they can distort.
That is why Agile teams need more than measurement: they need interpretation, context and honesty.
