Scheffers started Thunderstock in 2019. The company pairs a planning platform with merchandising specialists from retail, and handles budgets, collection planning and in-season stock management for fashion and shoe retailers across Europe. Sitting across a portfolio rather than inside one business, it sees patterns individual brands cannot.
He dates his own attention to the problem to a joke. Late July, the office in a heatwave, 30 degrees outside and a local water shortage, and a colleague pointed out that the winter money could finally start coming in.
The winter season begins at the end of July. What does that date refer to?
– Budget. A season code is a container in the accounts, not a description of a garment, and retailers plan with two seasons of roughly equal length because that is how the money is split. Meanwhile, because of climate change, the fashion winter season is shrinking rapidly. It sells later. Sometimes it does not sell at all, because winter never kicks in. Sometimes the peak arrives late and everything goes in a very short time.
Where does that cost them money?
– Winter goods start selling around the autumn holidays, so October. Black Friday arrives at the end of November, and Christmas follows immediately after. Once you have marked an item down you cannot put it back to full price in January, you are too late. So your window for full-price winter selling is October to Black Friday. In a typical winter that is workable. In a warm October and November it is not: customers wait, and you arrive at Black Friday holding the outerwear and the knitwear. From there you sell at markdown alongside everyone else.
Why are customers waiting?
– Buy now, wear now. Men were always quicker to buy for the weather in front of them, but we see it in womenswear and children’s clothing too. People used to buy the winter coat in midsummer to get the new collection before it sold out. Now they wait, because you cannot be sure winter will arrive at all, and if you spend a lot on a coat you barely wear, you have made a bad purchase. The weather does not explain all of it. Fast fashion is available across the year and drops are smaller and more frequent, which pushes the same way.
Which categories carry the risk?
– Outerwear and knitwear, because you cannot move them. Nobody introduces heavy knitwear in winter and keeps selling it through the summer as a summer article. The other direction is easier. T-shirts and shorts sell earlier and later than they used to, so they behave more like year-round goods: heavier materials for the colder months, lighter ones for high summer. A swimwear company told me recently that instead of marking down at the end of the season, they pull stock back from stores into online and keep selling at full price through the winter. There are always people going on holiday.
If the seasons are moving, what do the forecasting models get wrong?
– They smooth. A model leans on the most recent year but averages across several, so it understates extremes. It will never place a heatwave on a specific week, only raise the general probability. The second problem is the data. Retailers rebook articles between season codes: something sells through the summer, so at the end of the season they remap it to winter, and every system downstream has to be told. We do not assign those codes, our clients do. Which means you cannot read the season code as a statement about the product.
What is the first move for a brand planning against the wrong calendar?
– Keep a higher open-to-buy percentage going into winter, so the decision is made later. If the late summer holds, move part of that budget back into summer articles for August instead of putting coats and knitwear into stores. After that it becomes structural: more than two seasons, or drops instead of seasons, with sell-out calculated per drop. That is a large change for a business that has planned on summer, winter and never-out-of-stock for decades. And by the time the season is visibly wrong, they have already bought.