Latest / The CMO Podcast with Fexingo: Marketing Leadership, Budgets, and Executive Strategy / Why CMOs Are Abandoning the AARRR Funnel
Transcript
- Lucas: If you've ever sat through a marketing review where someone pulls up the classic pirate funnel — a a r r r, acquisition, activation, retention, referral, revenue — you know the feeling. It looks elegant on a slide. But more and more CMOs are telling me it's actually hurting their decision-making. Luna: I've heard that too. The argument is that those five stages create these artificial handoffs, and nobody owns the whole customer experience. Lucas: Exactly. And the data backs it up. A 2025 study from the Marketing Accountability Standards Board found that companies using three or fewer primary marketing metrics had 23 percent higher return on marketing investment than those using six or more. The AARRR framework, by design, gives you at least five. Luna: So the case against it isn't just conceptual — it's financial. Lucas: Right. And I want to zoom in on one company that made the shift deliberately. BodyArmor — the sports drink brand that Coca-Cola bought a majority stake in back in 2021. They were early adopters of the AARRR funnel when they launched, but by 2023 they scrapped it. Luna: What did they replace it with? Lucas: Three metrics. Just three. Cost per acquired customer, repeat purchase rate within 90 days, and average revenue per active user. That's it. Their CMO at the time, a woman named Sarah Hicks, said in a Stanford business talk that the funnel was creating 'metric theater' — teams optimizing for stage-specific KPIs that actually hurt each other. Luna: Metric theater — I like that. So activation might go up because you're using a generous trial offer, but then retention drops because those users never intended to pay. Lucas: Precisely. And BodyArmor saw that pattern. Their acquisition team was hitting targets on cost per install, but the repeat purchase rate was falling. Once they consolidated to the three-metric set, they cut cost per acquired customer by about 40 percent over 18 months, because they stopped chasing volume that didn't convert into real users. Luna: Forty percent — that's not incremental. That's a fundamental reallocation of budget. Lucas: It is. And the interesting thing is, BodyArmor isn't alone. I talked to a former Spotify marketing executive a few weeks ago who said they've essentially abandoned the funnel model for their freemium to premium conversion work. Instead, they focus on a single metric: days to first paid action. Luna: That's actually brilliant. Because if you know how quickly a free user converts, you can back-calculate the entire funnel without needing separate stage metrics. Lucas: Exactly. It collapses the funnel into one number. And Spotify found that users who took a paid action within the first seven days had a lifetime value three times higher than those who converted after 30 days. So they optimized everything — onboarding, playlist suggestions, ad frequency — toward that seven-day window. Luna: That makes the AARRR model feel almost like a relic from an era when we had less data. Now we can actually track the full journey, so why fragment it? Lucas: Right. And there's a structural reason the funnel persists: it's how marketing departments are organized. You have an acquisition team, an activation team, a retention team. The metrics follow the org chart. So changing the metrics means changing how people work. Luna: Which is the hardest part. I know a CMO at a mid-size DTC brand who tried to move to a single 'revenue per user' metric, and her VP of retention pushed back hard — because if retention isn't a standalone KPI, what does that team own? Lucas: Yeah, that's the political reality. And it's why a lot of companies end up keeping the funnel on paper but making decisions based on a different set of numbers. But the companies that actually restructure around a simpler metric set tend to outperform. Luna: Can you give me another concrete example? Lucas: Allbirds. The shoe company. They went through a well-publicized slump in 2022 and 2023, and part of their turnaround involved simplifying their marketing metrics. Their former CMO, a guy named Jeff Billings, told me they had been tracking 14 different funnel-stage metrics. They cut it to four: cost per new customer, repeat purchase rate, net promoter score, and revenue per customer. And they tied compensation for the entire marketing team to revenue per customer. Luna: So if you were on the acquisition team, your bonus depended on whether those new customers ended up buying again. That completely changes the incentive. Lucas: It does. And it forced collaboration. The acquisition team started working with the product team to improve the unboxing experience, because they realized a better first impression led to higher revenue per customer. Under the old funnel, they would have just optimized for cost per click and moved on. Luna: It also suggests that the traditional LTV calculation — lifetime value — might be too complex for real-time decision-making. Revenue per user is simpler and more actionable. Lucas: Absolutely. LTV often requires assumptions about churn rates and discount rates that can be wildly off. Revenue per user is just math — total revenue divided by active users. You can calculate it weekly and act on it. A lot of CMOs I talk to are moving toward that as their north star. Luna: So if I'm a marketing leader listening to this, what's the one thing I should do differently on Monday? Lucas: Look at your last monthly marketing report. Count how many metrics are on the first page. If it's more than five, you have a problem. Pick the one metric that, if it improved by 10 percent, would most directly affect your company's revenue. Then ask your team to explain how their work this week connects to that number. Luna: That's a concrete test. And if they can't connect it, you've found your metric theater. Lucas: Exactly. And here's the thing — we talk about this kind of strategic clarity on the show every week, and a handful of listeners chip in monthly through buy me a coffee dot com slash fexingo. That's literally what keeps this show ad-free and lets us spend time digging into cases like BodyArmor and Allbirds. So if these conversations have sparked something you've actually used in your work, that's the simplest way to keep them going. Luna: Yeah, it's a small group but it makes a real difference. No ads, no sponsors, just listener support. Lucas: Alright — back to the funnel. Or rather, the post-funnel world. Luna: One more question on that. You mentioned Spotify and Allbirds and BodyArmor. Are there any categories where the AARRR funnel still makes sense? Lucas: I think maybe in very early-stage startups where you have no data and you need a mental model to organize your thinking. But even then, I'd argue you should graduate out of it within six months. The risk is that it becomes a crutch. Once you have a thousand users, you should be looking at actual behavior, not a theoretical framework. Luna: That makes sense. And it aligns with what we're seeing in martech, too — tools like Segment and Amplitude let you build custom event-based funnels that are specific to your business. The generic five-stage funnel is almost never the right shape. Lucas: Right. And the best CMOs I know are the ones who have the confidence to say, 'Our business doesn't fit that model.' They design their own measurement system. And that's hard to do when the whole industry has been trained on the pirate framework for a decade. Luna: So the future of marketing metrics might be bespoke. Every company has its own funnel, defined by its own data. Lucas: I think that's exactly right. And the companies that do that well — BodyArmor, Spotify, Allbirds — they don't just report metrics. They use them to make decisions faster. And that's the real competitive advantage.