Have you ever wondered whether net promoter score (NPS) is really worth the effort? You’re not alone. In a recent study I ran, 82% of respondents said their company uses NPS. Yet only 34% said their CEO considers it “important” or “very important.” That disconnect says a lot.

So, is NPS a meaningful metric – or just a vanity number that looks good in a board deck? 

That’s the question I want to unpack.

Is NPS dead?

NPS has had its time in the spotlight, but skepticism is growing. Critics call it simplistic or even misleading.

Here’s my perspective: NPS isn’t dead, but the way many companies use it is. A single number won’t tell you everything you need to know about your customers, but if you design and run the program correctly, NPS can still be a powerful tool.

The key lies in how you structure it, how you analyze it, and – most importantly – what you do with the insights.

The good, the bad, and the ugly of NPS

What’s good about NPS

Two things make NPS appealing:

  1. It’s a single, simple question.

Every extra question you add to a survey cuts your response rate in half. Because NPS only asks one, you get more data points. That matters when you’re looking for statistically significant patterns.

  1. It’s benchmarkable.

Because the question wording is standardized, you can compare your score to competitors, different customer segments, or your own performance over time.

Where NPS falls short

Despite its popularity, NPS has serious weaknesses:

  • Weak correlation with retention. Research shows there’s little link between NPS and churn. If retention is your north star, NPS alone won’t get you there.
  • Limited actionability. A single score without context doesn’t explain why customers feel the way they do.
  • Low response rates. Many teams we surveyed reported just 1–10% participation. With data that thin, insights are shaky at best.
  • Inconsistent usage. Too often, teams tweak the question, embed it inside long surveys, or change the response scale. At that point, it’s not really NPS anymore.
  • One-off measurement. Sending a survey, publishing the score, and moving on doesn’t build a program. It’s just noise.
CTA Image

Ready for the full story?

Join Customer Success Collective as a Pro member for the full article, plus on-demand summit sessions, plug-and-play templates, and practical frameworks.

Sign up today

Seven ways to kill your NPS program

If you want to guarantee your NPS fails, here’s how:

  1. Embed it inside a long survey. You’ll kill the simplicity that makes it powerful.
  2. Change the question or scale. NPS works only if you keep the format consistent.
  3. Obsess over the score. It’s the trend that matters, not the number itself.
  4. Compensate employees on NPS. You’ll just incentivize gaming the system.
  5. Share only high-level results. Without detail, stakeholders won’t engage.
  6. Treat NPS as a health predictor. One data point can’t forecast churn.
  7. Fail to act on the feedback. If NPS lives in a dashboard, it’s wasted.