Marginally Better S01E18: The True Cost of Bad Customer Experience

What do a billion-dollar retail collapse, a broken app, and an abandoned shopping cart have in common? They’re all part of the same hidden bill, and most companies don’t realize they’re paying it until it’s too late.

In this episode of Marginally Better, Joe Taylor, Jr. breaks down the true cost of bad customer experience. Not as isolated mistakes, but as a slow, compounding erosion of trust that quietly drains revenue, loyalty, and brand value. From JCPenney’s infamous pricing overhaul to Domino’s brutally honest turnaround, this episode reveals a simple truth: the most expensive decisions businesses make are often the ones they never test—and the ones their customers never forgive.

Episode Links:

Transcript:

Joe Taylor Jr.: On February 1st, 2012, a man walked onto a stage in midtown Manhattan and told the country that American retailers had been getting it wrong for a hundred years.

His name was Ron Johnson. He’d just left Apple, where he’d built the most profitable retail stores per square foot in the history of the planet. Now, he intended to do the same thinsg for JCPenney. And on that morning in New York, in front of investors and reporters, he announced he was going to do it by getting rid of coupons.

No more sales. No more “40 percent off” stickers. Just honest, everyday prices. He called it “Fair and Square.” He smiled when he said it.

Eight weeks later, JCPenney’s same-store sales fell off a cliff. A year later, the company would lose nearly a billion dollars. And Johnson would be gone.

This week on the show — what did that hour on stage actually cost. Why the bill for a bad customer experience never arrives in one piece. And how a pizza company drowning in cardboard jokes did the one thing Ron Johnson refused to do.

That’s all coming up, after the break, on Marginally Better.

Announcer: From the global headquarters of Johns and Taylor in beautiful New Jersey, it’s Marginally Better. Here’s your host, Joe Taylor Jr.

Joe Taylor Jr.:
When we talk about bad customer experience, the conversation usually drifts toward stories. The angry tweet. The viral video. The friend who’ll never shop there again. Those stories are real. But they’re not where the money is hiding. The money is hiding in the spreadsheet.

Every year since 2016, the research firm Forrester has measured how American consumers feel about hundreds of major brands. They call it the Customer Experience Index. For most of the index’s history, the news was unremarkable — a few brands up, a few down, the average more or less flat.

Then 2024 happened. In their June 2024 report, Forrester analysts wrote that the index had hit its lowest point in nearly a decade. Thirty-nine percent of brands tracked saw their CX score decline year-over-year. Only nine percent improved. Senior analyst Rick Parrish told Fast Company that what they were watching was a slow leak — customers giving up on companies one disappointment at a time.

This isn’t a story about one bad year. It’s a story about the bill arriving for a decade of choices. Cheaper customer service. Slower email replies. Chatbots in front of humans. Websites built for the company, not the customer. Each of those decisions saved money on a quarterly report. But then, eventually, all of them showed up on the same line.

In May of 2024, the audio company Sonos pushed out a redesigned app. Sonos makes those wireless speakers – you might even have some in your kitchen. The app is how you control them.
The new version was, by the company’s own later admission, not finished. Features that customers had relied on for years were gone. Speakers stopped responding to commands. Reviews on the App Store collapsed all the way down to one star. Reddit lit up with people unable to play music in their own houses.

According to The Verge and Bloomberg, Sonos disclosed that fixing the app would cost between twenty and thirty million dollars in direct expense. The company also delayed two new product launches because the app couldn’t support them. Holiday sales suffered. By January of 2025, CEO Patrick Spence had stepped down.

Nobody at Sonos set out to break their own product. The app shipped because someone, somewhere, decided it was ready. The cost of being wrong about that — measured in revenue, in market value, in a CEO’s job — was almost certainly more than the cost of a few more weeks of testing.

If you’d like a number that explains why so many companies are bleeding without realizing it, here’s one. The Baymard Institute, which has been tracking online shopping behavior for over a decade, reports that the average e-commerce cart abandonment rate sits at right around seventy percent.

Seven out of ten people who put something in a digital cart never finish buying it. Some of that is window shopping. But Baymard’s research, which is built on usability testing rather than survey data, finds that the leading reasons are often fixable: surprise costs at checkout, accounts required when a guest checkout would do, slow loading, confusing forms. None of those problems require a new business model. They just require somebody whose job it is to notice.

That somebody, on most websites, sometimes doesn’t exist.

So here’s what these three stories have in common. The Forrester report tells us the experience is getting worse. The Sonos story tells us how fast a small mistake can compound. And the Baymard number tells us that most of the bleeding happens quietly, on a checkout page, without anyone setting off an alarm.

When we come back, the story of a man who walked into one of America’s oldest department stores believing he understood the customer better than the customer did — and what it cost to find out – for maybe the only time in his career to date – he was wrong.

That’s after the break, on Marginally Better.

Joe Taylor Jr.: It’s Marginally Better. I’m Joe Taylor Jr.

In the summer of 2011, the board of J.C. Penney was looking for a savior. The company was 109 years old. It had outlived two world wars and a depression. But by 2011, JCPenney was, by every meaningful measure, slowly drowning. Same-store sales had been flat or declining for years. The stores felt tired. The customer base was aging. Younger shoppers walked past the doors at the mall on the way to somewhere else.

Bill Ackman, the activist investor whose hedge fund Pershing Square had taken a major stake in the company, had a name in mind for the rescue. Ron Johnson. Johnson was a retail celebrity. He’d spent fifteen years at Target, where he’d helped invent the idea that a discount store could feel like a designer boutique. Then in 2000, Steve Jobs hired him to figure out what an Apple Store should be. By 2011, the answer to that question was thirty-eight hundred dollars in sales per square foot — the highest of any retailer in the United States.

Disclosure time — for a few of those years in the late aughts, Ron was my boss’s boss’s boss. I was part of Apple during a time when it had a notoriously flat org chart. We loved working for the guy. He could embrace getting a little cheesy during our all-hands presentations, and he wasn’t afraid to roll back an idea that wasn’t quite working right.
When Steve Jobs disclosed that his health issues were going to force him to step away from the business, many top level leaders there, like Ron, decided to go make their own dents in the universe outside Apple, just like Steve taught us.

So the pitch wrote itself. Take the man who had reinvented Target, then reinvented tech retail, and let him reinvent JCPenney. Johnson was named CEO in June 2011, brought a bunch of our former Apple colleagues over with him, and started in November.

He had, by his own count, exactly one hundred days to tell the world what he was going to do.

On February 1st, 2012, Ron stood on a stage in Manhattan in front of investors, journalists, and his own employees and unveiled a strategy he called “Fair and Square.” According to coverage at the time in the Wall Street Journal and the New York Times, the plan was bold in a way that retail rarely is. Johnson would eliminate nearly all of JCPenney’s roughly five hundred annual sales events. He’d get rid of the high-low pricing model the chain had used for decades — the model where everything has an inflated sticker price and then a coupon brings it back down to what people will actually pay.
In its place: three tiers. “Everyday” low prices. “Month-Long Values” on selected items. And “Best Prices” — the clearance rack, but renamed. No coupons. No flyers. No “40 percent off.”

Just honest pricing, all the time.

Onstage, Johnson explained why he thought this would work. Coupons, he said, were insulting. Sales events trained customers to wait until something went on sale. Real value shouldn’t need a red tag.

He also said, on the record, that JCPenney would not test these changes before rolling them out chain-wide. He told an audience at a Goldman Sachs conference, quote, “We didn’t test at Apple.” In Cupertino, that sentence had been a flex. At Apple Retail, change either iterated from the stores up – someone in the field would craft an interesting solution to the problem and it would bubble up until it became a global policy – OR, HQ would roll out a top-down transformation of a service, a feature, or a staffing approach.

Now, let’s take another look at that sentence, “We didn’t test at Apple.” In reality, we tested A LOT at Apple. There was a whole soundstage with a mock store inside of it meant to simulate what would happen if we ever changed one of the product displays, or where team members would stand, or adjusted the lighting temperature. Here’s what I think Ron was trying to say… With so much experience at the top of that flat org chart, few people argued when Ron wanted to experiment. Most of the things we did worked. The few that didn’t could usually be rolled back or forgotten fairly easily. Find one my colleagues from that time period and see if they’ll tell you why we all have a special relationship with the song “help me, Rhonda.” You see, this era was the pinnacle for Apple Retail’s growth. Riding the wave of iPhone, iPad, and MacBook Air, opening new stores at record pace — “We didn’t test at Apple.” because we were constantly trying to catch our breath.

In Plano, Texas, where JCPenney was headquartered, “We didn’t test at Apple” would turn out to be the most expensive sentence anyone had ever said inside the building.

The new pricing went live on February 1st. By the end of the first quarter — three months in — same-store sales were down nearly nineteen percent year-over-year. Johnson’s team assumed customers would learn the new system. They didn’t.

The Harvard Business School case study by Professor Joshua Margolis later documented what was actually happening. JCPenney’s core customer wasn’t shopping for the lowest possible price. They were shopping for the feeling of having found a deal. The coupon wasn’t an insult. The coupon was the fun! It was the point!

When you removed the coupon, you didn’t just remove a discount. You removed the dopamine. You removed the reason they came in the door. They didn’t show up on Saturday morning to buy a forty-dollar pair of jeans for forty dollars. She showed up to buy a sixty-dollar pair of jeans for forty dollars. Same transaction. Completely different experience.

By the second quarter, Johnson’s team started panicking. They added back some sale events. They reintroduced coupons under a new name. But the brand had spent a year telling customers that coupons were dead. Now the company was sending coupons. The message was incoherent. Same-store sales for fiscal year 2012 came in down 25 percent. The company lost nearly a billion dollars on the year. The stock, which had been near forty dollars when Johnson was hired, was trading in the teens.

On April 8th, 2013, the JCPenney board fired Ron Johnson. He’d been CEO for seventeen months. The board’s solution was to bring back his predecessor, Mike Ullman. Ullman immediately reinstated the coupons, the sale events, the flyers. He rolled back the in-store boutiques. He told employees the old playbook wasn’t broken. It had just been ignored. But the damage was done. JCPenney never recovered the customers it lost in 2012. The company filed for Chapter 11 bankruptcy in May of 2020. The chain still exists, in smaller form, under different ownership. But the version of JCPenney that Bill Ackman thought could be the next Apple — that company is gone.

Here’s the part of Ron’s story that doesn’t show up in the obituaries. Ron was NOT lazy. He wasn’t underprepared. He had spent his entire career studying retail, and he was, at the moment of his hiring, probably the most accomplished retail executive alive. The thing that took JCPenney down wasn’t ignorance. It was certainty.

He believed he knew, before any data could tell him otherwise, what the JCPenney customer wanted. He believed coupons were beneath them. He believed clean storefronts and Apple-style boutiques would speak louder than red sale tags. He believed all of this so completely that he refused to test it on a single store before rolling it out across eleven hundred locations. The cost of that certainty is now part of business school curriculum. A billion dollars in losses. Forty thousand jobs eventually lost when the company went bankrupt. What Forrester is measuring, in slow motion, across hundreds of brands — that’s the same thing on a smaller scale. It’s executives who are sure they know what their customers want, making decisions that nobody bothered to test, and then watching the bill arrive in pieces over the next few quarters.

The question I’d leave you with is the one Ron Johnson didn’t ask. How much of what you believe about your customer is really true? Is it something you’ve measured? And how much of it is something you’ve assumed?

After the break, an admission that could get me kicked out of New Jersey. It’s Marginally Better.

It’s Marginally Better, I’m Joe Taylor Jr.

In December of 2009, a different CEO of a different iconic American company stood in front of a different camera and said something almost nobody in his job has ever said out loud.

His name was Patrick Doyle. He’d just been named CEO of Domino’s Pizza. And in a four-minute commercial that the company released to the public, he said this, on camera: our pizza tastes like cardboard.

Not in those exact words. But close. The ad, which was the centerpiece of a campaign Domino’s called “The Pizza Turnaround,” showed real customers in focus groups saying real things. “Worst excuse for pizza I’ve ever had.” “The crust tastes like cardboard.” “Mass-produced, boring.”

Doyle didn’t argue. He sat in the room and took notes. Then he went back to his test kitchen, scrapped a pizza recipe the company had used since 1960, and rebuilt it from scratch — new sauce, new cheese, new crust.

And then he aired the commercial. The one with all the bad reviews in it. This is the move that probably should not have worked. Public companies do not, as a rule, broadcast their own customer hate mail. Marketing teams spend their lives keeping that footage out of the press. Doyle put it on national television and bought the airtime himself. The reaction was not what the marketing team expected.

According to coverage in Bloomberg and Advertising Age at the time, sales jumped fourteen percent in the first quarter after the campaign launched. Same-store sales kept rising for the next forty quarters. The stock went from under nine dollars in late 2009 to over four hundred dollars by the time Doyle stepped down in 2018. According to S&P data, that’s a return that beat Apple, Google, and Amazon over the same window.

The interesting thing about the Domino’s turnaround isn’t the new recipe. Plenty of companies change recipes. The interesting thing is what Doyle communicated by airing the bad reviews. He told customers, in language they couldn’t miss, that he’d been listening. Not in a focus group. Not in a survey nobody reads. Listening in the way that hurts — to the part where people say your product is bad and they’re not coming back.

And I say this as a person who moved to New Jersey, at least partially due to proximity to some of the world’s best fresh pizza, Domino’s pizza — got good!

There’s an episode of a Netflix food documentary with Chef David Chang that examines the whole phenomeon and urges viewers to take Domino’s for exactly what it is… not necessarily trying to replicate the world’s best pizza but making a solid product available in a lot of communities without access to a typical New York slice shop.

And even here in New Jersey, in a lot of towns like mine, where you suddenly find yourself binging episodes of Line of Duty after midnight and you need a pizza delivered, they’re you’re absolute best option.

Am I going to abandon my favorite pizziola who retired to Sicily then moved back to Jersey because he got bored? No way. But if I know I’ve only got a half hour delivery window and I’ve got a coupon in my digtal wallet, I am absolutely firing up that Domino’s pizza tracker.
Ron Johnson wouldn’t run a single test on one store. Patrick Doyle ran the focus-group tape on national television. One man was certain he understood the customer. The other one admitted he didn’t, and then went and found out.

The Forrester index, the Sonos app, the seventy percent of carts that get abandoned — all of those are what happens when a company is running on expert founder logic. They’ve decided what the customer wants. Nobody’s checking.

The Domino’s story is what happens when somebody finally checks. It is not, it turns out, the most expensive thing a company can do. It might be one of the cheapest.

The true cost of bad customer experience isn’t the refund. It’s the slow, quiet, year-after-year erosion of trust that doesn’t show up on any single quarter’s report — until, suddenly, it shows up on all of them.

If you’ve been listening to this episode and thinking about your own website, your own checkout flow, your own product, here’s the part where I’d usually pitch you something. So here it is, but softer than you’re used to.

A few times a year my team and I sit down with a single page on a single website, and we audit it the way Patrick Doyle’s team audited that pizza recipe. No assumptions. No “we already know.” Just: what’s actually happening here, what’s it costing, and what’s the smallest, cheapest thing we could change.

This year, I recorded an audio version of those audits and I called it the Website Reality Check. It’s at WebsiteRealityCheck.com and There’s a link in the show notes. In honor of the CEO that brought coupons back to JCPenney, use the promo code MIKE to take twenty dollars off your purchase. That’s at WebsiteRealityCheck.com.

Thanks for listening to Marginally Better. If something in this episode made you think differently, please leave us a review on Apple Podcasts or Spotify. It’s the cheapest possible way to help us and it genuinely helps the show find new listeners.

For show notes, transcripts, and the full source list for tonight’s episode, go to marginallybettershow.com.

Marginally Better is a Calufrax Radio production. Our producer is Nicole Hubbard. Research by Connie Evans. I’m Joe Taylor Jr. We’ll see you next week.

Marginally Better S01E18: The True Cost of Bad Customer Experience

Episode Links:

Transcript:

Marginally Better S01E18: The True Cost of Bad Customer Experience

Everybody’s Cindy Nowadays

Marginally Better S01E17: The Repair Relationship Revolution

Explore with me…