The Shelf You Give Up: Why Nike’s Direct-to-Consumer Bet Cost More Than It Saved

Ten months ago in this newsletter I called Nike the Toyota of shoes, a masterclass in industrial scale. That analysis still holds on manufacturing.

It was also incomplete, because while Nike was running the best supply chain in footwear, it was quietly dismantling the thing that actually carried its products to customers.

On 13 October 2024, John Donahoe retired as chief executive. The next day Elliott Hill took the job, a man who joined Nike as an intern in 1988 and had retired in 2020. He was brought back for one reason: to undo his predecessor’s central strategic bet.

This is the first piece in a new series. For two years this newsletter has published winners. The Moat That Broke looks at the other direction, because the mechanism that destroys an advantage teaches more than the one that builds it.

The Pain Point: Margin Looks Like Strategy

Every business with a distribution partner eventually runs the same calculation.

The retailer takes a cut. You do the manufacturing, the design, the marketing, the warranty. They put the box on a shelf and keep a share of the price. On a spreadsheet, cutting them out is the most obvious value creation available to you.

Go direct and you capture their margin, you own the customer relationship, you collect the data, you control the presentation. Every line of that argument is true.

The argument is also incomplete in a way that does not show up until it is expensive to reverse. You are not buying margin. You are buying an obligation to generate demand that the retailer was generating for you, at a scale you have never had to generate it before.

The Radical Story: Nike Left the Shelf

Under Donahoe, Nike pursued direct-to-consumer aggressively. Product was pulled from wholesale partners including Foot Locker, DSW and a long tail of independent retailers. The company pushed customers toward its own stores, its own website and its own apps.

The financial logic was sound. Direct sales carry the retailer’s margin. The customer data is yours. The brand presentation is controlled end to end rather than negotiated with a buyer who also stocks your competitors.

For a while it worked well enough to look like vision.

Then the shelves Nike had walked away from did not sit empty.

Retailers do not close the gap with apologies. They fill it. Foot Locker and every other partner needed product to sell, and competitors were willing to supply it. Running, in particular, was a category where specialist brands were building genuine credibility with exactly the customers who walk into a store to be fitted rather than ordering online.

By March 2024, Donahoe was publicly conceding that Nike needed to make important adjustments. That is a long way from where the strategy started.

Why it broke

The mistake was not going direct. Plenty of brands sell direct successfully.

The mistake was treating distribution as a cost line rather than as an asset that had been quietly doing work.

A wholesale partner is not just a margin-taker. They carry inventory risk. They employ staff who talk about your product to an undecided customer standing in front of three alternatives. They own physical locations in places you would never build. They generate demand for you, and the fee they charge is the price of that demand generation.

Cut them out and the invoice does not disappear. It moves onto your own income statement, as performance marketing, as retail leases, as customer acquisition cost. Nike swapped a variable cost it understood for a fixed cost it had to build from scratch, and it did so while handing its competitors the shelf space that had been keeping them small.

Then the ground shifted underneath. Consumer demand softened, and a business that had made itself responsible for generating all of its own demand had no partner absorbing any of the shock.

The Reversal, and What It Costs

Hill’s turnaround has been explicit about the direction: rebuild wholesale, reorganise the company around sports rather than consumer segments, and put innovation back at the centre.

The early evidence says the diagnosis was right. North America wholesale grew 11%, and running has now grown more than 20% for three consecutive quarters. Those are the exact areas the DTC strategy damaged.

But the reversal is not free, and the same results show what a rebuild actually costs. In the third quarter of fiscal 2026, revenue was 11.3 billion dollars, roughly flat as reported and down 3% currency-neutral. Net income fell 35% to around 520 million dollars. Nike Direct declined 4%. Greater China fell 17%.

Read those two paragraphs together and you have the whole lesson. The parts of the business being rebuilt are growing. The business as a whole is still paying for the years it spent dismantling them.

Getting back on a shelf is more expensive than staying on it, because you are now negotiating with a partner who has learned to live without you, and who has an alternative supplier sitting in the space you used to occupy.

The Part That Never Broke

There is a version of this story that treats Nike as a company that lost its way. That version is wrong, and the evidence is on the product side.

Through the entire distribution retreat, the innovation engine kept running. The Air Max 1000, the first Nike Air unit inside a fully 3D-printed shoe, did not quietly disappear when the strategy changed. It returned for Air Max Day on 26 March 2026 in a Black and Volt colourway at 179 dollars, followed by the 1000.2 in May with a refined lug design that is faster to produce, and then Air Works, a programme putting eight designers on their own printed Air Max styles running through to Air Max Day 2027.

That matters for the diagnosis. Nike’s product capability was never the problem. Its manufacturing was never the problem. The failure was confined to one decision about who carries the product to the customer, and it was severe enough to overwhelm everything the company was still doing well.

A single distribution error cost more than years of genuine innovation earned. That is how much distribution is worth.

The Playbook: Before You Cut the Middle Out

1. Price the demand generation, not just the margin

Ask what percentage of your sales come from customers who chose you at the point of sale rather than arriving decided. That number is what your partner is actually selling you. If it is high, their margin is not a leak, it is a fee for a service you would otherwise have to buy.

2. Assume the shelf gets filled

Model your competitor’s response, not just your own economics. Any distribution you vacate is capacity your rivals acquire cheaply. The relevant question is not what you save, it is what they gain and what it costs to take it back.

3. Recognise that you are converting variable cost into fixed cost

Wholesale margin scales down when sales fall. Retail leases and marketing teams do not. Going direct raises your operating leverage, which is excellent while demand grows and brutal when it does not. This is the same trade Ryanair runs in reverse, stripping fixed obligations out so the model survives a downturn.

4. Go direct in addition, not instead

The version that works is additive. Build the direct channel, prove it can generate its own demand at acceptable cost, and only then decide whether any partner is genuinely redundant. Nike had the strongest brand in its category and still could not carry the whole demand load alone.

5. Watch for the strategy that flatters the person proposing it

Direct-to-consumer arrives with better data, better control and better margin. It is the kind of plan that is easy to present and hard to argue against, and those are precisely the plans that need the most adversarial review before approval.

Nike did not fail at retail. It succeeded at building a direct channel and discovered, expensively, that owning the customer relationship is not the same as owning distribution.

The shelf you give up does not wait for you. Someone else is already standing on it.