Marketers know retention beats acquisition. They just keep dating new people.
The Beatles released You're Gonna Lose That Girl in 1965. The setup is two minutes long and brutal in its simplicity. The narrator is watching a boyfriend take his girlfriend for granted, and he's not warning the girl — he's warning the boyfriend. "If you don't take her out tonight / She's gonna change her mind." The girl already knows what's coming. The narrator already knows. The boyfriend is the only one in the room who hasn't figured it out, and by the time he does, she'll be gone, and there will be no fight about it. No conversation. No "we should talk." Just a quiet recalibration of who she's seeing on Saturday.
That's where most brands are right now with their best customers.
A friend told me at dinner tonight that they used to buy the same car brand, three cars in a row. Then the last one had a button fall off. Not a recall. Not a safety issue. A button. They didn't call the dealer. They didn't write a review. When the lease ended, they bought a different brand. The manufacturer has no idea why they lost a loyal customer, or at least, never made a sound.
The brand was the boyfriend in the song. The customer was already gone — emotionally — and nobody was watching. The actual switch was the easy part.
This isn't a failure of customer service. It's a failure of attention, and it's structural. Optimove's 2023 survey of 221 B2C marketing executives found that 54% allocate more than half of their marketing budget to new customer acquisition, while only 13% allocate more than half to retention. The 2022 numbers were nearly identical. Here's the brutal part: in the same survey, those marketers identified retention and churn prevention as the tactics that deliver the best ROI. They know where the money is. They just keep spending it somewhere else. The boyfriend knows the girl is unhappy. He's just busy planning a date with someone he hasn't met yet.
Now layer in what the academic research actually says about how customers leave. Marcel Zeelenberg and Rik Pieters published a regression study in the Journal of Business Research in 2004 that distinguished disappointment from generic dissatisfaction as an emotion with independent behavioral consequences. Disappointment, they found, is the strongest predictor of customer switching — stronger than dissatisfaction itself. It's also the strongest predictor of complaining and a strong predictor of word-of-mouth. The mechanism is specific: disappointment is what happens when delivery falls short of expectations, with the shortfall attributed to the brand rather than to the customer's own choices.
Here's the strategic problem that creates. Disappointment requires expectations to violate. First-time customers don't have specific expectations to violate — they have hopes, not yet contracts. Loyal customers do. The customers you've spent the most time building a relationship with are the customers with the highest expectations, which makes them the customers most vulnerable to disappointment-driven defection. The longer they've been with you, the higher the bar, the smaller the gap between expectation and reality has to be to register as a violation. A button falls off, and a 10-year relationship recalculates.
And the modern wrinkle: Zeelenberg and Pieters predicted in 2004 that disappointed customers would also complain — which gave brands a chance to intervene. In 2026, the cost of complaining has gone up (long hold times, AI menus, low expectation of resolution) and the cost of switching has gone down (one tap, abundant alternatives). The complaint step, which was historically the early warning, is being skipped. The customer goes straight from disappointed to gone, and the brand never knows there was a window.
So what do you do? Reallocate. The same Optimove survey shows marketers know retention is more profitable than acquisition; they need to actually spend like it. But the reallocation isn't just budget. It's attention. It's measuring expectation violation specifically, not just satisfaction. It's treating declining engagement among long-tenured customers as an emergency, not a curiosity. It's recognizing that the customer who used to buy four cars in a row is the one most likely to leave silently, and building a system designed to catch the silence before it becomes a switch.
This is the work The Joy Dividend is built around. Acquisition spend buys you attention. Keeping the customer requires something different — and the four pillars name what that something is. Calm Advantage is the floor: the button doesn't fall off, the post-purchase experience doesn't generate stress, the small frictions never accumulate into the breaking point. Surprise & Delight is what occasionally resets expectations upward instead of letting them quietly drift into "I expected better." Community & Connection is what makes a long-tenured customer visible to the brand — a customer who's part of something can't silently disappear, because the brand has eyes on them. Playful Design is the texture that keeps a relationship feeling alive rather than transactional. Together, the four pillars are the operating system for the retention budget marketers know they should be spending but aren't.
The narrator in the song doesn't go to the girl. He goes to the boyfriend. Because the girl knows. The girl always knows. The only person who needs the message is the one who thinks the relationship is fine because no one is complaining.
You're gonna lose that girl.
You should probably take her out tonight.
Originally published in The Joy Dividend on LinkedIn.