ANALYSIS · 10 JAN 2026

ZARA'S MÁLAGA STORE SELLS COATS IN AUGUST

Zara's store on Larios Street in Málaga sells coats in August because many of the people in it are Northern European tourists, who can buy a coat there at end-of-season prices before autumn starts at home. Pau Almar argues that demand planning should follow each store's actual customers, with weekly feedback from store data to buying decisions.

There is a Zara store on Larios Street in Málaga — one of the main pedestrian shopping streets in a city where the average August temperature is above 35 degrees Celsius. In August, that store sells coats. Not as a remnant from the previous season, not as a clearance operation, but as a deliberate commercial decision rooted in a reading of who walks through that door and why.

Most retail planners, shown this situation without context, would identify it as a mistake. Why would you allocate floor space to heavy outerwear in 35-degree heat? The answer shows how local demand planning works.

The Customer They Were Planning For

Málaga in August is not primarily a city for its residents. It is a city for its visitors — millions of tourists, a disproportionate number of them from Northern Europe, Scandinavia, and the UK. Many of them arrive on cruise ships for a few hours. Many of them are on the last day of a longer trip and haven't yet bought everything they came to buy.

A Norwegian tourist who has spent two weeks in Andalusia in late August has a specific relationship to coats. The ones available in Oslo at this time of year are priced for the beginning of the autumn season — full price. The ones available in Málaga's tourist-oriented shopping street are priced to sell, because the Spanish summer retail cycle has already moved into end-of-season territory. The same coat that costs 150 euros in Oslo in September is available for 89 euros in Málaga in August.

For a customer spending a day ashore from a cruise ship, this is an obvious purchase. They need the coat anyway. They have a few hours and a specific shopping mindset — looking for things they can take home rather than consume locally. The price difference is meaningful. The decision is not hard.

Zara's commercial teams understood this because they were reading demand across customer segments, not just across product categories. The question was not "what do people in Málaga buy in August?" The question was "what do the people who are actually in our store in Málaga in August want to buy?" Those are different populations with different purchasing behaviour.

Why Conventional Demand Planning Misses This

Standard demand planning operates on a calendar logic that is built around the average customer in the average location. Spring product arrives in stores in February and March. Summer product arrives in April and May. Autumn product begins arriving in August and September. The rhythm is driven by production lead times, buying calendar constraints, and a reasonable approximation of when the majority of customers in the majority of markets want each category.

This is a sensible system for managing inventory at scale. It works well on average. The problem with "on average" is that it is optimised for a customer who doesn't quite exist in any specific store.

Every physical retail location has a specific customer population — shaped by the neighbourhood it's in, the tourism patterns around it, the income profile of the catchment area, the proximity to transport hubs, the competition on the same street. These local factors create demand patterns that diverge from the national average in ways that the calendar-based planning system doesn't capture.

The same pattern appears in less obvious forms throughout any sizeable retail network. A store in a business district sells workwear at different proportions than a store in a residential neighbourhood three kilometres away. A store near a university has different size and price-point behaviour than a store in an upscale shopping centre. A store in a warm coastal climate needs a different category weight in its autumn collection than a store in an inland city 500 metres above sea level.

The brands that outperform at the store level are the ones whose planning systems capture and act on these local signals, rather than applying national-average logic uniformly.

The Data Behind the Intuition

Good demand planning in fashion is often described as part science and part intuition — the experienced buyer who "knows" that the Barcelona flagships will move product differently than the Madrid ones, without necessarily being able to articulate the model behind that judgment.

This is true, and the intuition part should not be dismissed. Pattern recognition built over years of watching specific stores perform is genuinely valuable. But the risk of relying primarily on intuition is that it is hard to transfer, hard to scale, and subject to the biases that individual experience creates.

The Zara approach is to make the intuition legible — to build data systems that capture the local signals that experienced merchandisers process informally, so that those signals can be acted on more consistently and at greater speed. When a store's sales data shows that a specific outerwear category is selling at twice the rate of comparable stores in the same climate region, that signal should trigger a reallocation of inventory toward that store before the buying team's quarterly review.

The operational requirement is a feedback loop between store performance data and buying decisions that runs faster than the traditional season review cycle. Not at the end of the season, when the learning comes too late for the current buying period. Not monthly, which is already an improvement but still too slow for fast-moving categories. Week by week, with enough granularity to see which categories are diverging from plan at which locations, and with buying decisions structured to respond.

Most fashion brands have the data to do this. Most of them review it too infrequently, or route it through too many layers of approval before it reaches someone who can act on it.

Season calendar and demand calendar

There is a retail assumption so embedded in the industry that it rarely gets examined: that the season calendar is a demand calendar. That customers want summer product when summer arrives and autumn product when autumn arrives.

This is true for most customers in most locations. It is false for enough customers in enough locations to matter materially to the P&L.

The customers who don't follow the calendar represent real demand that the standard planning model hasn't been built to serve. In some locations — tourist destinations, transport hubs, international shopping streets — these off-calendar demand signals represent the majority of the commercially significant traffic.

The data to respond to these signals exists. What varies is whether the organisation structures itself to use it — or continues to optimise for the median, predictable customer and leaves the rest to chance.

The principle behind the Málaga coats is to serve the customer who is actually in front of you, not the customer you planned for.

A Note on Scale

The relevance of this argument scales with the complexity of the retail network. A single-store brand doesn't need a sophisticated local demand system — the owner knows the customers. A brand operating 50 stores in five countries has a meaningful variation problem that its current planning system may not be capturing. A brand operating 500 stores across 20 markets has a local demand problem that is very probably costing it several percentage points of margin annually in missed sell-through and unnecessary markdowns.

The relationship between local demand responsiveness and full-price sell-through is consistent and significant. Brands that have invested in building it show better gross margins than comparable brands that haven't — not because of better buying at the aggregate level, but because of better allocation, faster reorder decisions, and a lower rate of inventory going to markdown.

The Málaga coats are a small line in a global operation; the same planning logic applies across the network.

// QUESTIONS

Why does Zara sell coats in its Málaga store in August?

Many shoppers on Larios Street in August are tourists from Northern Europe, Scandinavia and the UK. The same coat that costs 150 euros in Oslo in September is available for 89 euros in Málaga in August.

Why does conventional fashion demand planning miss local demand?

It follows a calendar logic built around the average customer in the average location, while each store has a specific customer population shaped by its neighbourhood, tourism patterns, catchment income, proximity to transport hubs and competition.

How much margin can weak local demand planning cost a fashion retailer?

For a brand operating 500 stores across 20 markets, the article says the local demand problem is very probably costing several percentage points of margin annually in missed sell-through and unnecessary markdowns.

← Back to Publications