Been tracking NPS for months but starting to question if we’re measuring the right thing.
Seeing high scores but retention tells a different story. Maybe the standard calculation misses something important about actual user behavior.
Anyone else noticed their NPS doesn’t match real loyalty metrics?
We break down our NPS by how much users spend and their session count. Paying users show completely different patterns than free users.
Same thing happened with our dating app. NPS looked great but users kept bailing after two weeks.
Turns out the high scores came from people who matched quickly. They felt good about the app but weren’t building real connections.
We split NPS by actual results - users who met someone vs. those who didn’t. Massive difference. People who never got past texting still gave us good ratings at first.
Now I weight responses by revenue and how often they use the app. A promoter who pays and opens it daily beats someone who rates you 9/10 then disappears.
Regular NPS treats everyone the same. Your business shouldn’t.
NPS is just a vanity metric. Track repeat purchases instead.
NPS measures intent but not actual behavior. That’s why you see high scores with low retention. Users might say they recommend your app but then uninstall it shortly after. The survey captures their mood, not their long-term value perception. Focus more on retention metrics when making decisions. A user who stays for months is more valuable than someone who rates you high but leaves quickly. To improve NPS, analyze scores by user tenure. New users often inflate their ratings.
High NPS scores can be misleading - they might come from users who barely touch your app but still feel good about it.
I always cross-reference NPS with how often people actually use the product. Daily users rate things way differently than weekly users. Survey timing screws with results too - ask someone right after onboarding and you’ll get artificially high scores.
Break down your NPS by user activity levels. That’s where you’ll see what’s really happening.