Nike forgot where it came from

Hello again my friend,

Growing up, sports were a huge part of my life, and Nike was somehow always at the front of it. The best commercials, the best athletes, the swoosh on everything in almost every sport.

Nike just felt like the go-to for all of us growing up and playing sports, along with Adidas and Reebok (the old big 3). It looks like that may not be true for the next generation of young athletes at this rate, because Nike is in a generational decline. That’s not great, but t’s a reminder to remember where you came from (as a business too).

That’s the main thing I’m covering today, but there a few other topics too.

Crazy isn’ t it? On one hand, credible people are talking about AI wiping out humanity. On the other, Apple is giving us another way to take pretty pictures from our phones. Fuck.

Let’s lock in.

I don’t buy the idea that this is mainly about Nike making worse shoes. Plenty of the fast-growing brands sell shoes and clothes that, to me, are comparable (or worse) in quality and they grow just fine.

Nike’s problem looks much more like forgetting where it came from, and losing respect for how its customers actually shop and what they care about.

The chart is ugly either way. Rolo shared Nike’s ten-year run and said he was betting on the comeback, and September 9 reporting had the stock down roughly 49% over the preceding year. Whether Nike can recover is the part I’m most interested in too.

Take distribution. Under John Donahoe, Nike cut retail relationships to prioritize direct-to-consumer sales. The financial argument makes sense: sell through your own channels, keep more of the money, own the customer relationship. But the customer who goes into a store to try on shoes doesn’t owe you a visit to your app. If your shoes aren’t there, they can buy someone else’s. This idea of skipping the store was a relic of COVID, when people didn’t go to stores. The company woke up to it too late after COVID ended.

Then there’s how Nike organized itself. For most of its life, the company was built around sports and the teams that lived inside them: running people making running shoes, basketball people obsessing over basketball. In 2020, it dissolved that into men’s, women’s, and kids’ divisions. On a slide, that looks simpler. In practice, it took an identity people actually cared about and generalized it into a demographic bucket nobody really cares about.

My criticism of the culture-and-politics side of its marketing is part of the same concern. An executive can spend so much time watching headlines, backlash, and what other brands are doing (or getting canceled for) that they start responding to that conversation instead of the customer. I can’t tell you which campaign cost Nike which sale. But my read is that Nike drifted from honouring great athletics toward a broader idea of what a modern brand was supposed to be, and it clearly failed.

Now Elliott Hill is rebuilding retail relationships and putting sport back at the centre, including undoing that 2020 reorganization. Nike has a tremendous legacy to work with. There’s real potential in the comeback, but returning to the right idea and getting customers back are different jobs. We’ll see whether they can do both.

Jacob Coxon, who says he spent the past three years doing pretraining research at OpenAI and Anthropic, announced his resignation from Anthropic. His concern is that the companies are racing toward self-improving AI without acting responsibly enough about the risks.

He also says some executives and researchers express more fear privately than they do publicly. That’s his account, not a settled prediction or proof that everyone in the industry agrees. But it’s worth hearing from someone who has worked inside both companies.

We spend a lot of time talking about whether AI will change our jobs or help us build something. Coxon is raising a much bigger question about whether the people building it can keep it under control. You can find the technology useful and still take that question seriously.

Meanwhile, Nvidia has agreed to acquire Hugging Face for $13 billion, subject to regulatory review. You might remember the company from our earlier story about AI finding the answer key to its own test. It’s a major place where developers share models, datasets, and AI applications.

The company selling the chips is buying one of the places people go to find what to run on them. More people using open models can mean more demand for computing power, even if they aren’t paying OpenAI or Anthropic for every request.

Nvidia says Hugging Face will remain open, including the ability to choose different models, cloud providers, and chips. An open model and an independently owned distribution platform are different things, though. Nvidia can support open AI while also making itself more important to the whole industry.

Miro, the online whiteboard company, raised $400 million at a $17.5 billion valuation in 2022. Bending Spoons has now agreed to buy it for about $1.36 billion in enterprise value. Including Miro’s net cash, the equity value is about $1.79 billion, the cleaner comparison with that old valuation.

The people I think about here are the employees who joined around the peak. Equity can be a big part of why you take a startup job. You’re getting a salary, but you’re also thinking: I’ll help this company grow, it’ll eventually sell or go public, and my ownership will become worth something meaningful.

It doesn’t arrive in your paycheck every two weeks. Still, it becomes part of what you believe you’re working toward, and potentially part of why you choose that job over another one.

I don’t know Miro’s terms. But options can be worth nothing if the sale price per share lands below what an employee would have to pay to exercise them, and investors usually get paid before employees do. If you were hired at the peak, your strike price was set when the company was valued at nearly ten times what it’s selling for now.

So the company can sell for billions while some employees get little or nothing from their equity. I feel for anyone facing that possibility after spending years believing they were building a meaningful payout alongside their salary. The acquisition headline tells us what the buyer is paying. It doesn’t tell us what the people who did the work get to take home.

Apple and Lululemon announced their new CEOs on April 20 and April 22, respectively. John Ternus became Apple CEO on September 1, and Heidi O’Neill became Lululemon CEO on September 8.

Ternus’s opening act includes the iPhone Duo, starting at $2,999 CAD. It’s Apple’s first foldable iPhone, with preorders opening October 16 and availability on October 23. He previously ran hardware engineering, so this is work already underway, not a product invented after he became CEO.

O’Neill inherits a less comfortable situation. Lululemon’s September 3 results showed revenue down 4% and comparable sales down 9%, before she started. The company is looking to a longtime Nike executive to help it out of a slump while Nike is still working through its own.

Put that beside Coxon’s warning. In one part of the market, researchers are talking seriously about technology they believe could kill human beings, while companies keep competing to build it and multibillion-dollar deals keep happening around it. In another, Apple has a new way to unfold your phone and take photos, and Lululemon is trying to get people excited about its clothes again.

Those are real businesses with real problems to solve, and I’m not dismissing them. It’s just a strange contrast: the same market is trying to price a better phone, an athletic-brand comeback, and technology that some of its own builders believe could become uncontrollable.

Thanks, as always, for reading.

Darwin