Adidas Never Actually Died. That's Why the Comeback Worked.

In October 2022, Adidas cut ties with Kanye West and inherited a warehouse problem: a pile of Yeezy sneakers it could no longer sell under the only name that made them worth anything. The line had brought in more than €1,200 million the year before.⁴ Overnight it became dead stock, a liability with a swoosh-shaped hole in the demand curve. The press did the math the way the press does — and declared a death watch. Adidas, the story went, was about to post its first loss in a generation and stagger back from the brink. It is a good story. It is mostly wrong.

The official version is that Adidas nearly died and clawed its way back. The truer version is quieter and more useful: Adidas was badly managed for a stretch, took a real hit, and course-corrected — but the company underneath the crisis never actually broke. It never posted an operating loss. The recovery worked precisely because there was far less to recover than the obituaries assumed.

The €700 million loss that never happened

Here is the number that drove the panic. In its 2022 annual report, Adidas told investors that 2023 could bring an operating loss of around €700 million — if it walked away from the Yeezy inventory entirely. That figure stacked up the worst case: roughly €500 million in lost Yeezy revenue, up to €500 million in inventory write-offs, and up to €200 million in restructuring one-offs.² It was a company saying, out loud, how bad things could get. Markets and headlines treated the warning as the forecast.

What actually happened was the opposite of a write-off. Adidas decided to sell the Yeezy stock anyway, donating a portion of proceeds to anti-hate charities including the Anti-Defamation League and quietly clearing the rest.¹⁰ Those sales generated roughly €750 million in net sales in 2023 and contributed about €300 million in operating profit.⁴ The feared liability became a salvage operation. Full-year 2023 operating profit landed at a positive €268 million — not a €700 million hole.¹ The single most-quoted catastrophe in the Adidas story is a guidance scenario that the company itself chose not to walk into.

A net loss the tax man wrote

But Adidas did post a net loss in 2023 — €58 million from continuing operations.¹ That is real, and it is the hook every 'near-death' retelling hangs on. So look at where it came from. The company turned an operating profit of €268 million into a net loss of €58 million, and the gap is almost entirely a 189.2% effective tax rate.¹ When your tax rate is nearly double your pre-tax income, the bottom line goes negative no matter how the business performed. This is an accounting artifact, not a structural failure — a company that earned an operating profit being pushed below zero by a one-year tax distortion. The Associated Press, reporting the result, called it a 'massive turnaround from net income of €254 million in 2022' and noted the company itself blamed the high tax rate.⁵

And the more famous figure — the €513 million loss that circulated widely — wasn't the 2023 result at all. It appears to derive from the heavily loss-making Q4 2022 period, a single bad quarter's figures retold as a full year of ruin. The math of the crisis got recycled, compounded, and inflated until a company that earned an operating profit two years running was being eulogized.

Why the comeback was easier than the legend

Björn Gulden took over as CEO on January 1, 2023⁹ and gets the credit for the recovery — the heritage pivot, the Sambas on every sidewalk, the wholesale repair. The recovery was real: by 2024, operating profit had jumped 398% to €1,337 million, revenue rose 11% to €23,683 million, and net income from continuing operations recovered to €824 million.³ Those are turnaround numbers. The question is what was actually turned around.

Gulden's own account is unusually honest about it. His strategy leaned on heritage silhouettes already in the catalog — Gazelle, Samba, Spezial, Campus, SL72 — plus localization, repairing the wholesale channel, and deliberately controlling supply to avoid over-distribution.⁷ Notice what's missing: an invention. The Samba nostalgia wave had been building organically — driven by celebrity sightings and social-media momentum — well before the turnaround strategy was formalised. He didn't conjure demand; he stopped suffocating it, then made sure Adidas could meet it without flooding the market and killing the scarcity that made the shoes cool. The turnaround mechanism wasn't a new brand. It was the discipline to stop over-shipping a brand customers already wanted.

But wasn't this still a genuine crisis?

The fair objection is that I'm downgrading a real emergency into a rounding error. Losing a partner worth more than €1,200 million in revenue overnight is not nothing.⁴ Fortune reported Adidas cut 2023 revenue guidance by around $1.3 billion and operating profit by roughly $536 million off the back of the Yeezy collapse, with an immediate estimated cost of $246 million.⁸ That is a brutal shock, and the stock and morale damage were genuine. So yes — crisis, full stop.

But severity and structure are different things, and the distinction is the whole point. Adidas has actually been near death before — a record loss in 1992 brought it close to bankruptcy, and it took Robert Louis-Dreyfus arriving in 1993 and an IPO in 1995 to rebuild it.⁶ That was a balance sheet at the edge. 2022–2023 was a profitable company that lost a celebrity, mismanaged its channels, and ate a one-year tax distortion. Both got called 'near-death.' Only one of them was. Confusing the two is how a course-correction gets sold as a resurrection.

The cleanest tell that Adidas was never on the brink is how the recovery happened. You cannot resurrect a brand that's dead — but you can stop strangling one that's merely been badly run. Gulden didn't perform CPR; he took his foot off the hose. The Sambas were already selling, the operating profit was already positive, and the catastrophe was mostly a tax rate and a headline. Adidas didn't come back from the grave. It came back from being managed as if it were already in one.