A KPI dashboard is a one-page view of the numbers a team has agreed to watch for the quarter. Metric tiles show the current value, the change against last quarter and a short trend. Goal bars show progress toward quarterly targets. A channel breakdown and a month-over-month table explain where the movement comes from, and a written key insight states what the numbers mean and where attention goes next.
Every KPI is a metric, and only a few metrics qualify as key performance indicators: the ones that tell people what to do to improve performance. David Parmenter separates key result indicators, which summarize outcomes for a board, from KPIs, which are measured often and point to a clear action. As a rough guide for a whole organization, he suggests about 10 key result indicators, up to 80 performance indicators and 10 KPIs. On a team dashboard, keep the tiles for numbers the team can act on this quarter and move the rest to a supporting table. Source: David Parmenter, Key Performance Indicators, 3rd edition (2015).
A lagging indicator reports a result after it has happened, and a leading indicator measures an activity or condition that drives that result. Revenue and churn are lagging. Activation rate, pipeline created and support response time are often leading, because they move first. Robert Kaplan and David Norton built the balanced scorecard on this pairing of outcome measures with the performance drivers that produce them. A useful dashboard shows both, so a team can act on the driver while there is still time to change the outcome. Source: Robert S. Kaplan and David P. Norton, The Balanced Scorecard (1996).
Show as many as fit on one screen without scrolling, which for most teams means a handful of headline numbers. Stephen Few defines a dashboard as a single-screen display of the most important information people need to do a job, designed to be monitored at a glance. Once a reader has to scroll, comparing one metric with another gets harder and the overall picture blurs. If the team cannot agree on what to cut, ask which numbers would change a decision this quarter. Keep those on the dashboard and move the others to a supporting report. Source: Stephen Few, Information Dashboard Design (2006).
Write two or three sentences that state what the numbers mean and what should happen next. Cole Nussbaumer Knaflic calls the core of this the big idea: a single sentence that carries your point of view and what is at stake. Lead with the result that changes the picture, name the gap that matters most, then say where attention goes in the next period. A reader who sees only the insight should still reach the right conclusion. Keep every number in it identical to the tiles, so nobody has to reconcile two versions. Source: Cole Nussbaumer Knaflic, Storytelling with Data (2015).
A KPI draws attention to what it measures, so people improve that number even when something else suffers. Andy Grove observed that indicators direct attention toward what they monitor, and recommended pairing each one with a counter-indicator that shows the side effect. If you track inventory reduction, also track stock shortages. If you track tickets closed, also track tickets reopened. On a dashboard, place each pair side by side, so a gain on one tile is read against its cost on the other. Source: Andrew S. Grove, High Output Management (1983).
| Metric | Month 1 | Month 2 | Month 3 | Trend | |
|---|---|---|---|---|---|
| Revenue | 88 | 94 | 102 | ↑ 15.9% | |
| Active Users | 9210 | 9840 | 10430 | ↑ 13.2% | |
| Avg. Session (sec) | 252 | 248 | 251 | → −0.4% | |
| Conversion Rate | 3.1 | 3.4 | 3.8 | ↑ 22.6% | |
| Support Tickets | 312 | 287 | 241 | ↑ −22.8% |