Latest / The Marketing Funnel Podcast with Fexingo: Awareness, Interest, Decision, and Action Stages / How NPS Scores Can Kill Your Marketing Funnel
Transcript
- Lucas: So there's this metric that almost every company with a CRM system tracks religiously. The Net Promoter Score. NPS. One question: 'How likely are you to recommend us on a scale of zero to ten?' Promoters are nine to ten. Passives are seven to eight. Detractors are zero to six. Simple, right? Luna: We've had NPS episodes on other Fexingo shows, but I don't think we've ever taken a knife to it on The Marketing Funnel. Lucas: Today we do. Because the more I dig into the research, the more I think NPS doesn't just fail to measure loyalty — it can actively break your marketing funnel. Specifically, it warps the Awareness and Decision stages. Luna: How so? Lucas: The metric was invented in 2003 by Fred Reichheld at Bain & Company. He wrote a famous Harvard Business Review article showing that companies with high NPS — like Enterprise rent a car, USAA, Harley-Davidson — grew faster than competitors. The logic was: promoters buy more, stay longer, and refer friends. So if you can optimize your funnel to produce promoters, you win. Luna: That sounds intuitive. Happy customers talk. Lucas: It does. But here's the problem. A 2024 study from Stanford's Graduate School of Business looked at a large software as a service company and tracked actual customer behavior versus NPS scores. They found that customers who scored nine or ten — the promoters — actually had the lowest lifetime value. They churned more, they used excessive support, and they were the most likely to demand discounts. Luna: Wait — the highest NPS customers were the least profitable? Lucas: Exactly. They were happy because they got a great deal. They gave a ten because the product was cheap. But cheap customers don't generate revenue. Meanwhile, the detractors — the fours and fives — were actually the most profitable segment. They paid full price, rarely contacted support, and stuck around for years. Luna: So the NPS was measuring satisfaction with the price, not loyalty to the product. Lucas: That's one reading. And it's devastating for marketing funnels because if you optimize for NPS, you start doing things that attract price-sensitive customers — heavy discounts, free trials, generous refund policies. Those tactics pull people into the Awareness stage, but they're the wrong people. They convert easily because the offer is too good, but they never become profitable. Luna: So the funnel fills with cheap leads. Lucas: Right. And then you track NPS as a success metric. The score goes up, everyone high-fives, but revenue per customer goes down. Tim Keiningham and his colleagues at the American Customer Satisfaction Index published a paper in 2018 — I think it was in the Journal of Marketing — where they analyzed 20 years of data and found that NPS explained less than one percent of variation in revenue growth across industries. Luna: One percent? That's basically noise. Lucas: It is. But companies still use NPS to set bonuses, to decide which marketing channels to invest in, and to judge campaign effectiveness. It creates a perverse incentive: make the experience so cheap that everyone rates you a ten, and then claim victory. Luna: I remember reading about a software company that literally fired its highest NPS customers. They analyzed the data and realized the promoters were costing them money, so they raised prices on them. Most left. The NPS dropped, but profitability soared. Lucas: That's the right move. But most marketers are terrified of a dropping NPS. It's a vanity metric. Social media likes are a vanity metric. NPS is a vanity metric dressed in academic clothing. Luna: Let's talk about how this warps the Decision stage specifically. Lucas: Sure. In a typical B2B funnel, the Decision stage is where a prospect chooses between you and a competitor. Many companies use NPS testimonials or case studies from promoters as social proof. But if those promoters are price-sensitive, the social proof attracts more price-sensitive prospects. It's a self-reinforcing loop that pulls your average customer value down. Luna: So the testimonials you put on your website are actively hurting your funnel. Lucas: If they're from the wrong promoters, yes. A better approach is to segment your NPS feedback by customer profitability. Ask: are our promoters actually our best customers? If not, stop using their quotes. Instead, use testimonials from customers who have high lifetime value, even if their NPS is lower. Luna: That requires actually measuring lifetime value per customer. Most companies don't. Lucas: They don't. They measure NPS because it's easy. One question. One number. But easy metrics lead to lazy decisions. Let me give you a concrete example from retail. A mid-sized clothing chain I studied — I'll call them 'Urban Threads' — had an NPS of 72, which is excellent. But their repeat purchase rate was declining. They did a deep dive and found that their highest NPS customers were one-time buyers who scored high because the returns process was frictionless. They loved the return policy, but they only bought once. The actual loyal customers were repeat buyers who scored lower because they sometimes had issues with sizing. Their NPS was 58, but they spent four times as much over two years. Luna: So Urban Threads was optimizing for return satisfaction, not purchase satisfaction. Lucas: Exactly. Their funnel was pulling in people who wanted easy returns, not people who wanted great clothes. The fix was to stop promoting the return policy in ads and instead emphasize fit guarantees and styling advice. They retargeted the low nps repeat buyers with a quality message. Within six months, their NPS dropped to 65, but revenue per customer increased 18 percent. Luna: That's a fantastic case. And it shows why blindly following NPS can lead you to optimize the wrong thing. Lucas: Look, I want to be fair to Reichheld. His original 2003 article made an important point: loyalty correlates with growth. But he was studying companies like Enterprise rent a car, where the promoter segment really was the best segment. In many modern digital businesses — especially subscription models and freemium — the dynamic is different. The cost of acquiring and serving a customer varies wildly. NPS doesn't capture that. Luna: What about the detractors? Should you ignore them? Lucas: No. But you have to ask: why are they detractors? Sometimes it's because they have a genuine problem with the product. Sometimes it's because they're power users who want more features. And sometimes it's because they just hate surveys. But here's the key insight from the Stanford study: the detractors who gave a score of six or lower but who contacted customer support and had their issue resolved actually became more loyal than customers who never had a problem at all. Luna: The service recovery paradox. Lucas: Exactly. That's a real phenomenon. So if your marketing funnel is structured to avoid detractors — say, by filtering out any prospect who seems skeptical — you're missing out on the people who, once converted, will become your best advocates after a good service experience. The funnel should welcome constructive friction, not avoid it. Luna: That's a radical idea for most marketing teams. They want everything smooth. Lucas: Smooth funnels produce average customers. Funnels that let some critics in and then serve them well produce raving fans. But you can't measure that with a single score. Luna: So if not NPS, what should marketers use? Lucas: I'd say three things. One: customer lifetime value segmented by acquisition channel. Two: actual retention rates and expansion revenue. Three: qualitative feedback that tells you why someone stays or leaves. The score is a proxy. The conversation is the data. Luna: You know, this conversation reminds me why we do this show without ads. The freedom to say NPS is overrated without upsetting a sponsor. Lucas: That's real. And honestly, if today's episode was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. It's how we keep this ad-free and honest. Luna: Yeah, it makes a difference. We don't have to sell you anything. We just get to talk about what actually works. Lucas: So back to the funnel — let's talk about how to apply this. If you manage a marketing funnel and you're using NPS as a north star, here's a practical step: run a cohort analysis. Look at the NPS of customers acquired through different channels. Then look at their 12-month value. I guarantee you'll find a channel where NPS is high but value is low. That's your discount channel. And a channel where NPS is low but value is high. That's your premium channel. Luna: Then you double down on the premium channel and fix the experience there. Lucas: Exactly. You don't try to raise the NPS of that channel. You try to understand what those customers need and serve it. The NPS may never go above 70, but your revenue per customer will climb. And your marketing funnel will start attracting the right people. Luna: What about the detractors in the premium channel? The ones giving a four? Lucas: Talk to them. Find out what's broken. Often it's a simple fix — a feature request, a billing issue, a communication gap. When you fix it, they become your biggest promoters. Not the nine-out-of-ten type. The kind who tell their colleagues in a one-on-one conversation. That's the most powerful marketing there is. Luna: And you can't buy that. Lucas: You can't. But you can design a funnel that encourages it. By not being afraid of detractors. By not optimizing for a vanity score. By actually listening to the people who care enough to complain. Luna: Alright, so next time someone on your team says 'our NPS is up,' ask them: up where? And at what cost? Lucas: That's the question. And if they can't answer it, you're probably marketing for the wrong customer.