Latest / The Marketing Funnel Podcast with Fexingo: Awareness, Interest, Decision, and Action Stages / How NPS Scored Can Kill Your Marketing Funnel
Transcript
- Lucas: If you run a marketing team, you've probably got a Net Promoter Score somewhere on your dashboard. One number — 'How likely are you to recommend us?' — and everyone from the CEO to the intern treats it like a report card. Luna: Oh yeah, NPS is practically a religion in some companies. I've seen teams celebrate a three-point quarterly increase like they just hit revenue targets. Lucas: Right, and that's the problem. Because NPS wasn't designed to be a marketing funnel metric. It was invented in 2003 by Fred Reichheld at Bain as a predictor of growth. But somewhere along the way, marketers started treating a high score as the goal itself — not what the score actually tells you about who's moving through your funnel. Luna: So you're saying a high NPS can be misleading? Lucas: Worse than misleading. It can actively warp your marketing strategy. Let me give you a concrete example. There's a B2B SaaS company — I won't name them because the data is from a private benchmarking study — that had an NPS of 92. That's elite. Apple-level territory. But when you looked at their actual referral traffic? Almost zero. Their customers loved the product, but they never told anyone. Luna: Huh. So the score said 'we're crushing it,' but the funnel was dead at the top. Lucas: Exactly. Meanwhile, there's another company in the same study — NPS of 31, which is considered poor — and they had a 40 percent month-over-month referral growth rate. Their customers weren't as effusive, but they were actually sending people. Luna: Interesting. So the low nps company was doing something right in the action stage, even if the sentiment was mediocre. Lucas: That's the key distinction. NPS measures sentiment intensity — how strongly you feel. It doesn't measure behavior. And the marketing funnel is ultimately about behavior: awareness, interest, decision, action. A promoter might be someone who just loves the brand but never shares it. A passive might be lukewarm but if you give them a simple referral link, they'll forward it to three colleagues. Luna: It reminds me of that classic stat — I think it's from a Google study — that people who rate a product 8 out of 10 are actually more likely to recommend it than people who rate it 10 out of 10. Because the 10s are often just happy users who don't feel the need to evangelize. Luna: I read that too. The 8s still feel a slight gap, so they talk about it to validate their own choice. Lucas: Yes. That behavioral quirk is exactly why blindly optimizing NPS can backfire. If you push every passive to become a promoter by over-delivering on service, you might actually reduce their likelihood to refer. They become so satisfied they stop thinking about the product altogether. It's the 'satisfied but silent' trap. Luna: So what should marketers do instead? Just ignore NPS? Lucas: No, not ignore — but restage it. The most useful way to think about NPS is as a diagnostic for the decision stage and the action stage, not as a top-line vanity metric. Here's the framework. Split your respondents into two groups: those who say they recommended you in the past 30 days, and those who didn't. Then look at their scores separately. Luna: So you're saying ask a second question: 'Did you actually refer someone?' Lucas: Exactly. And that second question is far more predictive of future growth than the raw NPS number. In the same SaaS study, the correlation between NPS and actual referrals was only 0.2 — basically negligible. But the correlation between 'did you refer someone in the past month' and next-quarter new customer acquisition was 0.7. Luna: That's a huge difference. So the action question is what matters for the funnel. Lucas: Right. And once you segment by behavior, you can target your marketing efforts. Instead of trying to turn passives into promoters — which might not even help — you focus on removing friction from the referral process for everyone. Make it easy to share, give them a reason, and then measure whether they actually do it. Luna: That ties directly to the bottom of the funnel. You're optimizing for action, not sentiment. Lucas: Absolutely. And here's where it gets really practical. A lot of companies run NPS surveys and then do nothing with the results except report the number. But if you add a simple follow-up — 'What prompted you to share or not share us recently?' — you start getting qualitative data that can inform your entire marketing funnel. You might find that promoters aren't sharing because they don't have a simple referral link, or because they're in an industry where word of mouth is rare. Luna: So you're basically using NPS as a starting point for a conversation, not as a final verdict. Lucas: Yes. And that's the mindset shift. The worst thing you can do is tie bonuses or team goals to a raw NPS target. That's how you get survey gaming, like calling customers to ask 'Would you give us a 10?' before they get the survey. Or only sending the survey to your happiest accounts. And then you're marketing to a distorted picture. Luna: I've seen that. One company I worked with had an NPS of 80, but when we actually looked at the raw response rate, it was only 5 percent. They were only surveying users who had just had a positive support interaction. Lucas: Classic selection bias. And that's why I think the most useful metric for the marketing funnel isn't NPS at all — it's referral rate. What percentage of your customers actually referred someone in the past quarter? That number, paired with a sentiment score, gives you a much clearer picture of whether your funnel is healthy at the action stage. Luna: So if referral rate is low but NPS is high, you have a referral friction problem, not a satisfaction problem. Lucas: Exactly. And if both are low, you have a product or experience problem. And if both are high, you're in a great place — but you still need to understand why people are referring, so you can double down on whatever's driving that behavior. Luna: Okay, so let's say a marketer is listening and wants to implement this. What's the first step? Lucas: First, stop optimizing for NPS alone. Second, add a behavioral question to your next survey: 'In the past 30 days, have you told anyone about us?' Third, segment your customers into four groups: high NPS + high referral, high NPS + low referral, low NPS + high referral, low NPS + low referral. Then design different interventions for each. Luna: I love that. It moves NPS from a vanity metric to an actionable funnel tool. Lucas: Right. And it aligns the marketing team with what actually drives growth — not just happy customers, but customers who act.