Many product teams have more dashboards than confidence.
They can see numbers moving, but still struggle to explain what the numbers mean, which behaviour sits underneath them, or what decision should follow. The dashboard may be correct and still be useless.
This usually happens because measurement started too late. The team began with the visible output — a chart, KPI, report, or stakeholder request — before agreeing what it needed to understand, which behaviour mattered, or what evidence would be good enough to act on.
That is the problem this playbook is trying to solve.
Dashboards cannot repair weak measurement
Dashboards feel like progress because they make data visible. They are familiar, easy to request, and easy to share. But visibility is not the same as understanding.
Two starting points
Dashboard-first versus measurement-first
Both routes may produce a dashboard. Only one begins by defining the understanding and evidence the team needs.
Dashboard-first
-
Dashboard request Begin with a report, KPI or stakeholder output.
-
Available data Use whatever events and fields already exist.
-
Charts and labels Arrange the data into familiar views.
-
Disputed meaning Discover that completion, engagement or success mean different things to different people.
-
Uncertain decision See movement without knowing what response is justified.
Measurement-first
-
Question or decision Define what the team needs to understand and why.
-
Workflow Identify the behaviour and boundary relevant to that purpose.
-
Observable evidence Decide what the product and team can responsibly observe.
-
Metrics and interpretation Calculate defined measures and assess their limitations.
-
Action or next question Use the evidence to decide, investigate, monitor or learn.
A team may see that registration completion has fallen without knowing whether the cause sits in the form, email verification, device behaviour, traffic quality, the tracking implementation, or the definition of completion itself.
The issue is not that dashboards are bad. A useful dashboard can help a team monitor, diagnose, evaluate, and decide. But it can only be as clear as the measurement thinking underneath it.
When the workflow, evidence, definitions, and decision context are weak, the dashboard becomes a place where uncertainty is displayed rather than resolved.
Start before the dashboard
A better measurement conversation starts with purpose.
What does the team need to understand? Which decision, investigation, or review would that understanding support? What are users or organisations trying to achieve? Where does that behaviour begin and end? What evidence would make the team more confident?
This does not mean every measurement exercise needs a long workshop or a perfect framework. It means the team should be able to trace a number backwards to observable behaviour and forwards to a reason for using it.
That traceability changes the questions teams ask.
Instead of:
What dashboard do we need?
Ask:
What do we need to understand, and what might we do differently once we understand it?
Instead of:
Which engagement metrics should we track?
Ask:
Which behaviour would show that people are receiving value, and what evidence could represent it honestly?
The harder questions come first. The charts come later.
What Measuring Products is for
Measuring Products is a practical guide to designing and operating product measurement systems.
Its central approach connects questions and decisions to workflows, observable behaviour, event evidence, metrics, interpretation and the next action.
The playbook explains how to map meaningful workflows, define useful event evidence, calculate trustworthy metrics, design dashboards around decisions, and keep the whole system healthy as the product changes.
It is not a guide to a particular analytics platform, a catalogue of generic KPIs, or an argument for collecting more data. Tools can record, calculate, and display. They cannot decide which behaviour matters or whether a metric deserves to influence a product decision.
That judgement belongs to the team.
The purpose of product measurement is not to produce more dashboards. It is to understand product behaviour well enough to make better decisions — and to know how much confidence the available evidence deserves.