What Is a KPI Dashboard and What Should Yours Show?

The word 'dashboard' gets used to mean almost anything these days. A tab in Excel with some charts. A page in a report. A screen full of numbers. But a real KPI dashboard is something more specific — and more useful.

A KPI dashboard is a single screen that shows the metrics leadership needs to understand how the business is performing right now, compared to targets and trends, without any manual preparation.

What a KPI is — and what it isn't

A KPI (Key Performance Indicator) is a metric that is directly linked to a business objective. 'Total sales this month' is a metric. 'Monthly recurring revenue as a percentage of target' is a KPI.

The distinction matters because a dashboard full of metrics without context doesn't help anyone make a decision. A dashboard with KPIs — defined relative to targets, compared to prior periods, flagged when they move outside expected ranges — does.

• A good KPI has a clear definition that everyone agrees on
• It is calculated consistently from the same data source every time
• It has a target or benchmark to compare against
• Someone in the business is accountable for it

How many KPIs should a dashboard show?

The most common mistake in dashboard design is showing too many metrics. When everything is highlighted, nothing is. A senior leader reviewing a dashboard should be able to answer 'how are we doing?' in under 30 seconds.

Most businesses need 6–10 KPIs on a primary dashboard. Finance-focused businesses might need a few more. Operational dashboards for delivery teams can go slightly higher. But if you find yourself listing 25 KPIs, you have not prioritised — you have postponed the decision.

What a good KPI dashboard structure looks like

1. Summary row at the top

Three to five headline numbers — the most important KPIs with their current value, target, and a simple up/down indicator. Anyone glancing at the dashboard can see status at a glance.

2. Trend charts below

Month-on-month or week-on-week trend lines for the key metrics. This is where you see whether a number is improving or declining over time — context that a single figure can't give you.

3. Breakdown views

Split the headline numbers by the dimensions that matter most — by product, by region, by team, by customer segment. This is where analysis starts.

4. Exception flags

Automatic callouts for anything that has moved significantly — outside target range, below a threshold, or showing an unusual spike. The dashboard should surface the things that need attention, not require the reader to hunt for them.

The most important step before building anything

Before opening Power BI or Tableau, write down the exact definition of each KPI. How is revenue calculated — does it include VAT? Does it count orders placed or orders invoiced? What counts as an active customer?

These definitions seem obvious until you ask three people and get three different answers. Agreeing them first — in writing — is the most valuable thing you can do before a dashboard build starts.