Not subscribed? Sign up to get it in your inbox every week.

Starbucks optimized itself into irrelevance. Your company might be doing the same thing.

I know three agency owners who crossed $1 million in revenue last year. Combined, they paid themselves less than a mid-level project manager at any of their clients.

Same story every time. They'd figured out growth landed bigger accounts, hired more people, added service lines. The dashboards looked great. Revenue up and to the right. Utilization rates that would make a consultant blush. And somewhere along the way, each of them reached out separately with the same question: Where did the money go?

It went into the machine. They scaled the business. They also scaled away the value.

More headcount to service the new accounts. More tools to manage the new headcount. More overhead to support the tools. They built a revenue engine that consumed everything it generated including the thing that got them their first clients: the owner doing exceptional work.

I've been calling it the curse of scale: when your operation executes so well that it optimizes away the thing that made the metrics possible. The expensive version of it I've ever seen played out at Starbucks over seven years with the entire business world watching and nobody saying a word.

February 14, 2007. Valentine's Day. Howard Schultz sits down and writes a memo to his CEO, Jim Donald. Subject line: "The Commoditization of the Starbucks Experience."

Starbucks has 13,000 stores. Revenue has more than doubled in four years $4.1 billion to $9.4 billion. The growth engine is performing exactly as designed.

And the man who built this coffee slinging behemoth from a single Seattle store has just written down that they've destroyed it in the process.

The memo shows up in every business book about brand dilution. But most treat it as a story about losing your way. That's the wrong frame. Starbucks didn't lose anything. They executed perfectly and grew their way into the problem.

Between 2000 and 2007, Starbucks made three operational decisions. Each one was correct by the metrics it was judged against. Each one solved a real constraint. Together, they gutted the brand experience of a coffee lovers haven in exchange for sugared smoothies and soccer moms.

The machine swap. Starbucks replaced its fleet of semi-automatic La Marzocco Lineas with Thermoplan Verismo 801 super-automatics. The La Marzocco sat 14–17 inches tall on the counter. You watched the barista grind, dose, tamp, and pull. The Verismo stood 24–28 inches integrated grinder and bean hopper on top. A wall between the barista and the customer.

Push-button espresso eliminated 30 seconds of active barista time per shot. A few hundred espresso drinks per store per day times 13,000 stores and the labor/training savings alone ran north of $100 million a year. With turnover in the double digits annually and tens of thousands of new hires cycling through, reducing the skill floor for espresso was the only way to survive at scale.

Nobody measured “romance and theatre.”

The bag switch. Open bins of whole beans scooped and ground in front of customers replaced by flavor-locked, pre-ground packaging. Consistent product. Longer shelf life. Faster service.

No loose coffee beans = no rich beany aroma. Schultz called it "perhaps the most powerful non-verbal signal we had in our stores." Quality and consistency were up again, and they lost the coffee shop lovers for the commuter crowd.

The cookie-cutter buildout. Arthur Rubinfeld, who'd built Starbucks' original design philosophy sending teams into neighborhoods, sourcing local materials, designing stores that felt indigenous to their locations left in 2002. After his departure: modular design kits, templated floor plans, corporate art. Comfortable seating replaced by harder furniture that encouraged faster table turns.

Once again scale took its toll. You cannot custom-design seven stores a day.

One sentence told the whole story:

"Many of these decisions were probably right at the time, and on their own merit would not have created the dilution of the experience; but in this case, the sum is much greater and, unfortunately, much more damaging than the individual pieces."

Three rational decisions. No villain or disaster or upstart competitor. Just a machine doing exactly what it was built to do while quietly suffocating the core of the brand experience.

My agency friends each fell into their own versions of this curse. One replaced her personal client onboarding and strategy calls with a templated workflow. One hired “rockstar strategists” with their own styles to handle the deliverables she used to sign off on herself. One discounted his biggest proposals to boost AOV and scale. None of them were wrong. All of them were killed profits.

The Starbucks scorecard during this entire period was financial and volume-based. Same-store sales. New store openings. Revenue. Operating margin. EPS. The numbers looked great.

But same-store sales YoY comps are made of two things: transaction count and average ticket. Starbucks reported strong comps every quarter. But transaction growth had dropped for four straight years. Fewer people walking through the door every single year. The headline comp held because average ticket kept climbing price increases, food attach, larger sizes.

Making more per visit masked that visits were disappearing.

FY2008: same store sales of -3%, driven by transaction declines of 5–6%. The signals had been sitting in the publicly reported data for years.

By 2002, internal research showed a "highly satisfied" customer visited 7.2 times per month lifetime value of roughly $3,170. A merely "satisfied" customer: 3.9 visits, lifetime value $920.

That's a 3.4x gap between someone who loved the experience and someone who tolerated it.

Repeat visit frequency wasn’t on the scorecard. Same with espresso quality, "Third place" perception, aroma, theater. The feeling that you were somewhere that truly loved coffee. The things that justified a premium went unmeasured and unprotected as they were slowly eroded away by the necessities of scale.

January 2008 Schultz comes back as CEO. Stock down 42%. Same-store sales declining for the first time in company history.

Seven weeks later, he closes all 7,100 U.S. stores simultaneously for three and half hours of "Espresso Excellence Training." Signs on the doors: "We're taking time to perfect our espresso."

Dunkin' ran specials outside shuttered locations. Analysts called it a stunt. Schultz, in Onward: "…it would be perceived as our own admission that Starbucks was no longer good enough. But if I was honest with myself, I knew that that was the truth."

The retraining was the start of the surgery. 600-plus store closures. 6,700 layoffs. More than $330 million in restructuring. It took nearly three years to stabilize.

But Starbucks didn't go back to semi-automatic machines. They worked with Thermoplan to develop a next-generation super-automatic designed with a deliberately lower profile to restore sightlines while keeping push-button efficiency. They redesigned the automation to protect the experience.

You can't unscale. The answer is: figure out what your scaling mechanism is silently destroying, and then redesign the mechanism so it stops.

I asked a fellow COO “How do you know your customer experience is excellent?”. He thought for thirty seconds and responded: "The number of clients who mention a specific person's name in their renewal call." That's his aroma. When that number drops, the experience is dying no matter what the utilization dashboard says.

Howard Schultz built the company and signed off on every one of its biggest mistakes along the way. He defined "the romance of coffee" after optimizing it out of existence. He just didn't see it happening until the damage was done, and wrote a GOATed memo to begin the repairs.

Searching for your next role? Fill this form out, and we’ll intro you to the best companies in the world.

Hiring? just respond to this email! We’ve got dozens of vetted operators standing by.