Most businesses pour their energy into two kinds of customers: the brand-new ones they’re trying to win, and the superfans who’d follow them anywhere. The trouble is that neither group is where the business actually lives. The big, unglamorous majority in the middle — the regulars who already signed up and keep paying without making a sound — is the group almost nobody designs for, and the group most likely to drift away without a word.
In this episode of Marginally Better, Joe Taylor, Jr. makes the case for the forgotten middle: the gym-economics study that found the most profitable members are the ones who barely show up, the loyalty programs built for VIPs while the rank-and-file go dormant, and the Microsoft Office team that discovered customers were begging for features the software already had — then tore the whole interface down to serve the silent majority. Plus a no-cost assignment for finding your own forgotten middle this week.
Episode Links:
The Gym: Paying Not to Go
- DellaVigna & Malmendier — “Paying Not to Go to the Gym,” American Economic Review (2006)
- Full working-paper PDF (UC Berkeley)
Loyalty Programs: The Forgotten Member
- Bond Brand Loyalty — The Loyalty Report 2023
- McKinsey — “Next in Loyalty: Eight Levers to Turn Customers Into Fans”
The Design Principle: Perpetual Intermediates
The Office Ribbon: Jensen Harris
- Jensen Harris — “The Story of the Ribbon”
- Jensen Harris — “No Distaste for Paste (Why the UI, Part 7)”
- “The Most Frequently Used Features in Microsoft Office”
- Harvard Business Review — “Why Microsoft Had to Destroy Word”
Product Mentioned
Transcript:
[00:00:00] Somewhere this month, a person is going to pay eighty dollars for a gym they walk into four times. They’re not the muscle-bound regular who lives on the squat rack. And they’re not the New Year’s resolution that’s already gone. They’re the one in the middle — the member who means well, shows up now and then, and pays full freight for the privilege. And here’s the strange part: that member, the one barely using the place, is the most profitable customer the gym has.
Now hold that thought, because it’s not really about gyms. It’s about a blind spot that runs through almost every business on earth.
We spend enormous energy on two kinds of customers. The brand-new ones — we obsess over the welcome, the onboarding, the first five minutes. And the superfans — the power users, the top tier, the people who’d tattoo our logo on their arm. But between those two groups sits everybody else. The big, unglamorous majority who already signed up, already know us, and are slowly drifting toward the door while we look the other way.
So today: the gym that profits from the people who don’t come. The loyalty card in your wallet you forgot you had. And the software team that discovered its customers were begging — for features the product already had. That’s all coming up, on Marginally Better.
Every business sorts its customers, whether it means to or not. There are the newcomers, the regulars, and the die-hards. And almost everywhere you look, the attention flows to the two ends — the people just arriving and the people at the very top. The enormous group in the middle gets treated like furniture. It’s always there, so we stop seeing it. The trouble is, that’s usually where the money is. And it’s where the leaks are too.
Start at the gym, because two economists put hard numbers on it. Stefano DellaVigna and Ulrike Malmendier published a study in the American Economic Review back in 2006 with a title that says it all: “Paying Not to Go to the Gym.” They got the actual records from three health clubs — about seventy-seven hundred members — and matched what people paid against how often they actually walked in.
And the members who picked the flat monthly plan, the one running over seventy dollars a month, went on average about four times. Four. Which works out to more than seventeen dollars every time they swiped in — at a gym that sold a ten-visit pass for ten bucks a visit. They could have paid less by the punch card and they chose the membership instead. And they hung on to it. Those monthly members were markedly more likely to still be paying a year later than the people who’d signed an annual contract, because canceling means admitting something you’d rather not admit. So the gym’s best customer isn’t the regular sweating through a workout. It’s the hopeful person in the middle, paying every month for the version of themselves they keep meaning to become. The question is whether the business even knows that person is its bread and butter.
Now reach into your own wallet, or your phone. The Bond Loyalty Report, which surveys this every year, found the average American belongs to around eighteen loyalty programs — and is active in only about half of them. So right now you are a “member” of nine or ten things you’ve effectively forgotten. Enrolled. Counted in somebody’s quarterly deck. Doing nothing.
And here’s what the companies do with that. McKinsey, looking at how these programs are built, points out that loyalty schemes are designed around the very best customers — the top-tier flyer with the VIP line, the white-glove treatment, the invitation to the special event. Meanwhile the ordinary member in the middle gets a points balance they’ll never spend. Which would almost make sense, except McKinsey’s own numbers show where the upside hides: an active member spends about ten percent more than one who’s enrolled but checked out — and the ones who actually redeem rewards spend twenty-five percent more. The growth isn’t at the very top, where you’re already maxed out. It’s in waking up the middle. And almost nobody’s pointing the budget there.
So why does this keep happening? The interaction designer Alan Cooper gave the cleanest answer I’ve found, in his book “About Face.” Line up the users of almost any product by skill, he says, and you get a bell curve. A few rank beginners on one end. A few true experts on the other. And this giant bulge in the middle — people who are competent, comfortable, and not going anywhere. Cooper calls them “perpetual intermediates,” because — and this is the part that stings — they mostly stay there. Beginners don’t stay beginners long. Experts are rare and hard to keep. The middle is permanent, and it’s most of everybody.
But look at who ends up designing for whom. Cooper’s point is that the managers tend to fixate on beginners, because that’s who they picture when they imagine a customer. And the engineers build for experts, because that’s who they are. Nobody in the room is the intermediate. So the biggest group of all gets designed for by accident, if at all. Which brings me to a company that fell into exactly that trap — and then did something almost nobody has the nerve to do. It tore the whole thing down.
When we come back: a software team sitting on a mountain of customer complaints, all asking for the same thing — and the slow, awful realization that the thing people were begging for had been in the product the entire time. They just couldn’t find it. What that team decided to do about it became the way you use a computer today.
Quick word about something we made, because it lands right on top of today’s whole idea. Here’s the thing about your own website. You can’t see it like a stranger anymore. You built it, or you’ve stared at it for years, so you navigate it like an expert — you know where everything is. And the people who built it with you are experts too. But the person you actually need to reach is that intermediate in the middle. Someone who’s heard of you, maybe visited once or twice, knows roughly what you do — and is now trying to get one specific thing done before they give up and close the tab. Nobody on your team experiences the site the way that person does. That’s the blind spot.
That’s exactly why we built the Website Reality Check. It’s a plain-spoken, second set of expert eyes on your site — we go through it the way that middle visitor actually would, and we tell you where they get stuck, where they doubt you, and where they give up and leave. No jargon, no fifty-page audit you’ll never read. Just a clear read on what your site is really like to use, and the handful of fixes that’ll move the needle. If you’ve ever had the feeling that your site is fine for you but you’re not sure it’s fine for them — that’s the feeling this was built for. You can find it at johnsandtaylor.com, under Services. Website Reality Check.
It’s Marginally Better. I’m Joe Taylor Jr.
Let’s go back to the early two-thousands, to a team inside Microsoft with a problem most companies would kill to have. Their product was a runaway success. It was also, by their own admission, becoming a monster.
The product was Microsoft Office — Word, Excel, PowerPoint. And the man I want you to follow is Jensen Harris, who ran the Office user-experience team. Picture his world for a second. Word had started life back in the late eighties as a fairly simple thing — a blank page and a few menus. But every couple of years, Microsoft added more. More features, more buttons, more toolbars. By 2003, Word could put roughly thirty different toolbars on your screen. There was a paperclip named Clippy whose entire job was to help you find the features you already owned and couldn’t locate. The program had everything. And that turned out to be the problem.
Because here’s what was landing on Harris’s desk. Customer requests. Thousands of them, pouring in, asking Microsoft to please, please add certain features to Office. And when the team actually read through the pile, they noticed something that stopped them cold. The features people were demanding — the things they swore were missing — were, over and over, already in the product. Already built. Already shipping. Sitting right there under a menu nobody opened. People weren’t asking for new power. They were asking for a flashlight.
Now, Harris’s team had something most designers of that era only dreamed about. Real data. Office had a program — the Customer Experience Improvement Program — where users could opt in and let the software report which commands they actually used. Not what they said in a survey. What they really did, millions of sessions of it.
And the data told a brutal, clarifying story. In Word, the single most-used command was Paste. Just Paste — about eleven percent of everything people did. The top five commands — Paste, Save, Copy, Undo, and Bold — added up to about a third of all the clicking in the entire program. Paste was number one in Excel too, and in PowerPoint. And after the first ten commands or so, the graph just collapsed into a long, flat tail of hundreds of features that almost no one ever touched. By Harris’s own account, something like ninety percent of users were living inside less than ten percent of the product.
So sit in his chair. You’ve got a program with hundreds of features. You’ve got a customer base begging for things the program can already do. And you’ve got a giant middle of users — not beginners, not experts, just regular competent people — who plateaued years ago and never went deeper, because every new feature got buried under the last one. The obvious move, the safe move, is to keep doing what you’ve always done. Add the next feature. Ship the next toolbar. Keep the power users happy and paper over the rest with another helpful paperclip.
Harris’s team made the other choice. They decided the interface itself was the thing that was broken — and they were going to tear it down to the studs. No more menus. No more stacks of toolbars. They’d build something that put the right tools in front of you at the moment you needed them, surfaced the features people had been blindly asking for, and was designed, deliberately, for that competent person in the middle. They prototyped obsessively — Harris later said the team generated so many design variations that when he went to give a talk about it, he had something like twenty-five thousand screenshots to pick through. The thing they landed on, they called the Ribbon.
Office 2007 shipped with the Ribbon. And the first thing that happened was that a lot of people got furious.
Because here’s the cruel math of designing for the middle. The power users — the small, loud group who’d memorized every menu — woke up one morning and found their furniture moved. Everything they knew was gone. And they let Microsoft have it. If you go back and read the comments on Harris’s own blog from that time, they’re something. “Worst idea ever.” People announcing they were ripping it out and rolling back to the old version. Whole offices in revolt. The experts, the ones who’d mastered the old way, experienced the change as a punishment. And they were the most vocal customers Microsoft had.
It would have been so easy to lose nerve right there. The loudest voices in the room are almost always the experts, because they’re the ones with the most invested in how things already work. If you only listened to the volume, you’d put the menus right back.
But the bet wasn’t on the loud end of the curve. It was on the silent middle — the millions who never wrote a comment, never memorized a menu, and now, for the first time, could actually see what their software could do. And that bet held. The Ribbon didn’t just survive the backlash. It became the template. Look at the program you’re probably using to read your email, or write a document, or build a deck, right now — the strip of tabs and tools across the top. That’s the Ribbon, roughly twenty years on, copied across the whole industry. Microsoft took the heat from the experts to serve the people in the middle who’d never say a word. And the people in the middle, it turned out, were almost everyone.
Which leaves a question worth carrying out of here. When your loudest customers and your most numerous customers want opposite things — and they will — whose voice are you actually building for?
I want to leave you with something you can do this week, because the good news in all of this is that the forgotten middle isn’t a problem. It’s an opportunity sitting in plain sight, and almost nobody’s working it.
Think about where your attention naturally goes as a business. You celebrate the new customer — the fresh signup, the first sale. There’s a little dopamine hit in it, and you build whole funnels to chase it. And you cherish the superfan — the one who refers everybody, leaves the glowing review, shows up to everything. They’re a joy. But ask yourself, honestly, when was the last time you thought hard about the customer who’s been with you eighteen months, uses you out of habit, has never complained, and has never raved either? The one who’s perfectly satisfied and one mild annoyance away from drifting to a competitor without ever telling you why?
That person is your business. Not the edges — the middle. And Alan Cooper’s insight from earlier is the key to the whole thing: that customer isn’t a beginner you need to hand-hold, and they’re not an expert you need to impress. They’re a competent grown-up who wants to get something done without friction. They don’t need more features. Like the Office users begging for tools they already owned, they probably can’t even use half of what you offer now. What they need is for the thing they already do with you to get a little easier, a little clearer, a little more respectful of their time.
So here’s the assignment, and it costs nothing. Pick your middle. The customers who’ve been around a while and use you regularly and rarely make a sound. Then go find out what mildly annoys them — the small recurring friction they’ve decided to live with because complaining isn’t worth the trouble. You won’t find it in your five-star reviews and you won’t find it in your churn report, because these people haven’t left yet. You have to go ask. And when you fix that one small thing for the quiet majority, you don’t win a headline. You win something better. You win the people who were always going to decide your future — and never planned on telling you so.
That’s our show. If you take one thing with you this week, take this: your business doesn’t live or die at the edges. It lives in the middle — with the regulars who’d never call themselves fans, and would be hard to replace if they left. Go see them. They’ve been waiting.
And if you want to start with the place that middle customer meets you most often — your website — that’s exactly what we built the Website Reality Check for. We look at your site the way the person in the middle actually does, and we tell you the truth about where they get stuck. You know where to find us.
Thanks for listening to Marginally Better. If something here was useful, do me a favor and pass it to one person who runs a business — that’s how this show grows, one forwarded episode at a time. And leave us a review while you’re at it — it’s the single best way to help one more business owner find the show. For show notes and the research behind today’s episode, head to marginallybettershow.com, or just follow the link wherever you’re listening.
Marginally Better is a Calufrax production. Our producer is Nicole Hubbard, with research by Connie Evans. I’m Joe Taylor Jr.
