Counts lie, rates tell the truth

The cleanest way to decide what stays on your dashboard comes from Eric Ries, and it is worth stating plainly because it settles most arguments. A vanity metric measures the size of your business but tells you nothing about what to do next. An actionable metric links a specific change to an observed result: we changed X, and Y moved. That distinction is the entire case for measuring rates over counts, and once you internalise it you stop defending half the charts you used to keep.

The one-line test

For every metric on your board, ask one question: if this number moved, would I know what to do differently? If a number can swing up or down and leave you with no clear action, it is a vanity count and it comes off the decision dashboard. It can live on a report somewhere, because reports are for describing the business to other people. Your decision dashboard is for steering it, and steering needs metrics that tie a cause to an effect.

Lead count fails the test. If leads jump 30 percent, what do you do? You do not know, because the count says nothing about whether those leads are good, whether they convert, or where they stall. Website traffic fails it too, for the same reason. A conversion rate, by contrast, passes cleanly. If your MQL-to-SQL rate drops from 35 percent to 28 percent, you know exactly where to look and roughly what is wrong. You changed nothing upstream and a downstream rate moved, which means the problem lives at that joint.

Leading and lagging

The second cut to make is between leading and lagging indicators, because confusing the two is how teams end up steering by the rear-view mirror. Lagging metrics confirm what already happened. Closed revenue and net revenue retention are lagging, they tell you the truth but they tell it too late to change the quarter that produced them. Leading metrics move this week and predict the lagging ones. Stage conversion rates and speed-to-lead are leading, they shift in days, and a shift in them today shows up in closed revenue weeks later.

You need both, but you steer with the leading ones. A founder who only watches closed revenue is watching a number they can no longer influence by the time they see it move. A founder who watches the stage conversion rates and speed-to-lead is watching the levers that will produce next quarter's revenue, while there is still time to pull them. The lagging numbers are your scoreboard. The leading rates are your controls.

So the work of cleaning a dashboard is mostly subtraction. Run every chart through the one-line test, bin the counts that fail it, and sort what survives into leading and lagging. What you are left with is small, and small is the point, because a metric you can act on beats ten you can only stare at.