Ries defines vanity metrics as numbers that go up but do not tell you whether your business is actually working. Total registered users is the classic example. A thousand people signing up sounds great. But if nine hundred of them never return after the first day, the sign-up number is hiding a retention disaster. Actionable metrics, by contrast, are directly connected to specific decisions. They tell you whether something you changed made a meaningful difference. The test of whether a metric is actionable is simple: if the number changes, does it tell you what to do differently? If not, it is probably vanity. Ries recommends what he calls cohort analysis: instead of looking at aggregate totals, track groups of customers who came in at the same time and measure how their behaviour changes over time. This reveals whether improvements are actually improving the right things, or just masking problems with volume. One company Ries worked with was celebrating total revenue growth month after month while their customer retention was quietly getting worse. The vanity metric looked great. The actionable metric was flashing red. They nearly missed it.