The Challenge
In the outerwear category of one buying season, the proposed range included three quilted jackets priced between €149 and €179. They came from different brands, with minor differences in lining weight and zipper detail. Each had a legitimate case for inclusion when reviewed individually. Taken together, they were competing for the same shopper at the same price point for the same occasion, and none of them was going to capture enough of that demand to sell through at full price.
A mid-sized European multi-brand fashion retailer carried ranges structured like this across outerwear, knitwear, and trousers. Buyers had selected products carefully. Brand mix was intentional. Category reviews signed off on every SKU. By week ten of the season, markdowns had already begun. Not on a handful of obvious slow movers, but broadly across the range.
The pattern repeated every cycle. Full-price sell-through peaked early, then stalled. Markdown depth increased quarter over quarter to clear the tail. The buying team attributed the pattern to shifting consumer demand. The finance team pointed to inventory depth. Neither explanation was wrong, but neither identified the root cause: assortment overlap.
Overlap across outerwear, knitwear, and trousers, the range carried multiple products occupying the same functional and aesthetic space: similar silhouettes, adjacent price points, overlapping colour families, and comparable fabrication. Together, they split demand across more SKUs than any of them could achieve critical velocity on. Products that might have reached full-price sell-through rates of 75 to 80 percent were instead reaching 45 to 55 percent, because demand was distributed too thinly across a range with too many near-identical options.
This is a structural problem across the industry. More than 40% of fashion goods are eventually sold at markdown, and the share of assortments on discount increased by 5 percentage points in the first half of 2025 compared to the prior year.
The cost is significant: Markdowns compress gross margin directly and they also train a segment of the customers to wait for discounts, which suppresses full-price conversion in subsequent seasons and makes the next markdown cycle structurally necessary.
A structural problem across the industry
Assortment overlap is not a judgment failure: buyers cannot hold the full competitive interaction of a 300-SKU range in their heads simultaneously. It is a data problem, and one that compounds over seasons as carryover products are renewed into ranges without systematic review of how they interact with the new buy. Running this analysis continuously, at machine speed, converts pre-season planning from a review exercise into a precision allocation decision. More than 40% of fashion goods are eventually sold at markdown, and the share of assortments on discount increased by 5 percentage points in the first half of 2025 compared to the prior year.
The Solution
The retailer deployed Zenline's Assortment Agent in a pre-season capacity, running overlap detection across the proposed buy before purchase orders were finalised. The agent analysed the incoming range against four dimensions: silhouette and product category, price point clustering, colour family and fabrication, and target shopper occasion derived from existing sell-through data by product type.
What the Agent did:
- Mapped the full proposed buy into functional clusters by shopper occasion and price tier, scoring each cluster by overlap intensity and projected markdown risk based on historical velocity data for similar product configurations
- Flagged 18% of proposed SKUs as high-overlap candidates: products where demand was likely to be cannibalised by a stronger alternative already in the range
- Prioritised the flagged list by commercial impact, presenting buyers with a ranked view of which consolidations would have the largest effect on projected sell-through
- Ran the same analysis on carryover SKUs to identify products being renewed into a range that had already demonstrated overlap-driven underperformance
Examples, what it found:
- Across knitwear and trousers, overlap-adjusted projected sell-through for flagged SKUs was 38%, versus 71% for the non-overlapping range
- Carryover SKUs flagged as overlap candidates had underperformed full-price sell-through targets by an average of 22 percentage points in the prior season
The Impact
Margin and sell-through:
- Full-price sell-through in the pilot categories increased by 14 percentage points versus the prior season's equivalent
- Average markdown depth in cleared products fell from 34% to 21%, as fewer items reached end-of-season without prior velocity
- Gross margin in the pilot categories improved by approximately 2.1 percentage points season over season
Buying process:
- Open-to-buy redirected from cut SKUs was reinvested in depth on the three best-performing silhouettes per category: products that subsequently sold through at 84% full price
- Pre-season review time for the categories in scope fell by roughly 30%, as buyers received a prioritised overlap report rather than reviewing 300+ SKUs individually
- Markdown approval meetings for the pilot categories were reduced from three in-season events to one, as clearance volume was substantially lower
Organisational:
- The buying team moved from a reactive markdown management posture to a pre-emptive one: the shift in cadence allowed earlier, shallower price interventions when they were needed, rather than deeper markdowns under time pressure at season end
"We had always known the range was too similar in places. What we didn't have was a way to see it clearly before we committed the budget. The agent gave us a ranked list of what to cut and why, with the sell-through projections behind it. The first season we used it, we had our strongest full-price performance in four years."
— VP of Category Management, European Multi-Brand Fashion Retailer