The death of the dashboard

For twenty years, when a product team did not know what a screen should say, they shipped a dashboard. It was the polite compromise between the team that had the data and the team that had the users. It let us feel generous without committing to a point of view. It is time to admit that this compromise is over.

What the dashboard was really for

A dashboard is a promise dressed as a product. The promise is: somewhere inside these tiles is the number that will change your Monday. The user's job is to find it. This is a strange arrangement — we built the tools, we know the schema, and yet we hand the reader a wall of charts and wish them luck.

"A dashboard is what you ship when you are afraid to have an opinion in public."

The thing that replaces it

What replaces the dashboard is not a better dashboard. It is a product with a stance. A single sentence at the top of the screen that says: here is what changed since yesterday, here is what we think you should do about it, here is the one action worth taking. Everything else is evidence, arranged in support of that sentence.

• 01Lead with a claim, not a chart.
• 02Rank the decisions, not the metrics.
• 03Make the second screen the place where you go to argue with the first.

Why we don't do this

Because opinions can be wrong, and dashboards can't. If we say "churn is up because onboarding is broken" and we are wrong, someone can point at us on Friday. If we say "here are 42 charts, you decide" and the customer churns, that is the customer's fault. Dashboards are a liability shield disguised as generosity. Killing the dashboard means taking the liability back.

What you get back

In exchange for the risk, you get a product people can recommend in a sentence. Nobody has ever recommended a dashboard. They have recommended tools that told them what to do next.