The State of Fashion 2026, published by The Business of Fashion with McKinsey, sets out what an elevating brand is supposed to do with its channels. The instruction is one line: protect brand equity by carefully restricting distribution through outlets and off-price channels, particularly for signature products. Deloitte's Global Powers of Luxury Goods 2026 shows the same reflex further down the profit and loss account, with 39.3 percent of surveyed executives expecting store closures or network optimisation, rising to 57.3 percent in apparel and footwear.
Nobody in the industry disputes this. Three houses have now run it at scale and published the results quarter by quarter, and only one of them got what the theory promised.
Restricting distribution and withdrawing from it look identical in a board paper. They are not the same move, and the difference has nothing to do with the quality of the plan.
At Nike's investor day on 25 October 2017, Elliott Hill put the number on the record. He was then president of geographies and integrated marketplace, and he is now chief executive. Nike was doing business with 30,000 retailers across 110,000 points of distribution, and would execute its direct strategy with about forty partners. North America would move from 40 percent differentiated retail to 80 percent in five years. Trevor Edwards, then president of the Nike brand, named the remainder on the same stage: retail that does not offer quality service or powerful storytelling is undifferentiated retail, and Nike would be shifting away from it.
It worked, in the sense that the plan did what it said. Nike Direct rose from roughly 28 percent of Nike brand revenue in fiscal 2017 to about 44 percent in fiscal 2023 and 2024. Gross margin peaked at 46.0 percent in fiscal 2022.
Then, on the second-quarter call of 19 December 2024, Hill described the position he had inherited. Some partners and channels, he told investors, feel we turned our back on them, and we stopped engaging consistently. On the same call, chief financial officer Matt Friend was more precise about the mechanism: Nike Digital had become a platform where the company was capturing demand and competing with its own wholesale partners rather than creating demand for its brands. Friend then named the price of going back. Nike would fund higher wholesale discounts, in his words, to win back shelf space.
Gross margin in fiscal 2025 was 42.7 percent, the lowest of the arc, and Nike's own filing attributes the fall to higher discounts, changes in channel mix and higher inventory obsolescence reserves. By June 2025 Hill was describing a footprint of 40,000 points of distribution. The same story ran in miniature on a single platform: Nike ended its Amazon pilot in November 2019 to focus on more direct, personal relationships, and on the first-quarter call of 30 September 2025 Hill reported that the new Nike brand store on Amazon was driving stronger engagement and sales than anticipated.
The shelf was not handed back. It was repurchased, at a discount, from the retailers who had been told they were undifferentiated.
Foot Locker's own annual filings track the vacancy on one fixture. Nike accounted for 75 percent of all merchandise Foot Locker purchased in fiscal 2020. By fiscal 2024 it was 59 percent. Sixteen points of a major retailer's buying book changed hands, and the retailer had every reason to help it happen: a buyer who has been told his shop does not deserve the best product will diversify on his own account, and he will not undo that plan when the supplier comes back.
McKinsey's Sporting Goods 2025 report for the World Federation of the Sporting Goods Industry says plainly who took it. When Adidas and Nike pivoted toward direct-to-consumer channels in the late 2010s, the report finds, challengers stepped into the vacuum; they adopted wholesale-first strategies and teamed up with specialist retailers to gain credibility and expand their footprint.
Absence is not scarcity. Absence is a gap in a planogram, and planograms get filled.
On Holding is the clearest case, because its filings refuse the flattering story. On has been a wholesale-majority business in every year since its 2021 listing; wholesale was 58.2 percent of net sales in 2025 and direct has never exceeded 42 percent. In that same year On reported a gross margin of 62.8 percent, its highest ever, and attributed it to operational efficiency and the strength of its premium positioning rather than to channel mix.
| Measure | Period | Value |
|---|---|---|
| Nike points of distribution | 2017 | 110,000 |
| Nike points of distribution | 2025 | 40,000 |
| Nike share of Foot Locker purchases | FY2018 | 66% |
| FY2019 | 71% | |
| FY2020 | 75% | |
| FY2021 | 68% | |
| FY2022 | 65% | |
| FY2023 | 65% | |
| FY2024 | 59% |
The highest gross margin in this set belongs to the house that never left the shelf.
Dr. Martens moved its direct share of revenue from 49.3 percent in the year to March 2022 to 64.8 percent by the year to March 2025, and wholesale revenue fell from £479.6 million to £276.9 million across the same period. Gross margin behaved exactly as the textbook says it should, rising from 63.7 to 65.0 percent. Operating margin fell from 25.2 percent to 4.7 percent.
On 5 June 2025 the company published its own account of why. A focus on a direct-first approach, it wrote, led to a loss of coverage and responsiveness in the wholesale offering, restricting growth and reducing consumer touch points, which in turn produced reduced customer acquisition, elevated inventory and increased use of clearance channels, particularly in the United States.
The same document closes the comfortable escape route. Dr. Martens states that its brand and product range are not over-distributed, either across wholesale accounts or in terms of price architecture. The problem was never too many doors. It was too few, held too tightly, while the cost base stayed sized for the business the company used to be.
| Fiscal year | Gross margin | Operating margin | Direct share |
|---|---|---|---|
| FY22 | 63.7% | 25.2% | 49.3% |
| FY23 | 61.8% | 17.6% | 52.1% |
| FY24 | 65.6% | 13.9% | 60.8% |
| FY25 | 65.0% | 4.7% | 64.8% |
| FY26 | 66.2% | 7.5% | 62.9% |
A gross margin that rises while operating margin falls is not elevation. It is the same business absorbing the retailer's costs and reporting the result as a premium.
Birkenstock's 2023 listing prospectus uses the phrase engineered distribution twenty-seven times, and the mechanism sits in one sentence the company has repeated in its filings ever since. It generates significantly more demand from existing and prospective wholesale customers than it can supply, which puts it in a position to create scarcity in the market and obtain consistently favourable economic terms on wholesale distribution. Allocation runs, in the company's own words, often down to the single door level. In fiscal 2022 average selling price rose 25 percent while units sold rose 3 percent, and Birkenstock attributes part of that to exiting certain distribution and wholesale partnerships in key markets.
The honest part of the case is the part nobody quotes. In fiscal 2025 Birkenstock's direct share fell from 40 to 38 percent, because business-to-business grew 20 percent against direct's 11, and the company's own margin bridge lists channel mix as a drag on the year. Engineered distribution is not direct-to-consumer maximisation. It never was. It is the right to say no, and that right is created upstream, in supply, not downstream, in a channel strategy.
The test takes ten minutes and belongs in the next range review. Before a door is cut, write two sentences. The first names what the next brand on that fixture will do with the space. The second names the thing this house is holding back that the door cannot order from anybody else. If the second sentence comes out empty, the decision on the table is not a tightening of distribution, and it should not be presented to a board as one.
A house does not own its shelf by leaving it. It owns the shelf by being the one thing on it that cannot be reordered from somebody else.
Double opt-in. Six emails a year, nothing else, unsubscribe in every one. Privacy.
A confirmation email is on its way to . Signals starts once you click the link inside it.