Every quarter, somewhere in a boardroom, a slide goes up.
NPS: 42.
The room nods. Someone writes it down. The CEO says "good, let's keep that up." The meeting moves on.
Nobody asks what happened to the customers who gave you a 6. Nobody asks why the promoters are not actually referring anyone. Nobody asks what a 42 means compared to last quarter, or what it means for revenue, or what anyone is supposed to do differently on Monday morning because of it.
NPS is the metric that makes everyone feel like they are measuring customer health without having to do anything uncomfortable about it.
What NPS actually measures.
Net Promoter Score asks one question: on a scale of 0 to 10, how likely are you to recommend us to a friend or colleague?
Promoters are 9s and 10s. Passives are 7s and 8s. Detractors are 0 through 6. Your score is promoters minus detractors as a percentage.
The problems with this are numerous.
First, it measures intent, not behaviour. A customer who says they would recommend you is not the same as a customer who actually does. The correlation between high NPS and actual referral behaviour is weaker than the consultants selling NPS software would like you to believe.
Second, the scale is arbitrary and the interpretation varies by industry. An NPS of 42 is excellent in telecommunications and mediocre in consumer software. Comparing your score to last quarter without industry context is noise dressed up as signal.
Third, it is a lagging indicator. By the time your NPS drops, the damage is done. The customers who were going to churn already decided. The score is the obituary, not the diagnosis.
Fourth, and most importantly, it does not tell you what to fix. A detractor gave you a 4. Why? You do not know. The survey did not ask. You have a number and no information.
What to measure instead.
Three metrics that actually connect to revenue and actually tell you what to do:
Churn rate by cohort. Not overall churn. Churn by acquisition channel, by onboarding path, by product tier, by customer size. When you break churn into cohorts you stop seeing a blended average and start seeing patterns. The customers from that one campaign in Q3 churn at three times the rate of everyone else. That is actionable. That is a budget decision. That is a conversation with sales about the promises being made.
Time to value. How long does it take a new customer to experience the thing that made them buy? This is the single most predictive metric for long-term retention I have encountered across twenty years of growth work. Customers who hit their first value moment quickly stay. Customers who do not, leave. Measure it. Shorten it. Everything else follows.
Expansion revenue rate. What percentage of your revenue growth comes from existing customers expanding versus new customers being acquired? Companies with high expansion revenue have fundamentally different unit economics than companies that rely on new acquisition for growth. If you do not know this number you do not know whether your product is actually delivering on its promise.
None of these fit on a single slide as cleanly as NPS: 42.
That is the point. The metric that fits cleanly on a slide is usually the metric that requires the least accountability from everyone in the room.
The metrics that tell you what to do are harder to present, harder to explain, and harder to argue with.
Start there.
Rob
P.S. If you are going to keep running NPS surveys, at minimum add one follow-up question: "What is the primary reason for your score?" The open text will tell you more than the number ever will. Read every response yourself. Do not delegate this.