By Elie Dufeu, CTO & Co-Founder, Metreecs. Published 1 July 2026.
The fashion industry generates approximately 92 million tonnes of textile waste every year. Of that total, an estimated 4-9% of unsold textiles placed on the EU market are destroyed before ever being worn, representing between 264,000 and 594,000 tonnes of incinerated or landfilled textiles per year in Europe alone (Global Fashion Agenda; European Commission).
That figure now has a regulatory consequence. The EU's Ecodesign for Sustainable Products Regulation (ESPR) bans the destruction of unsold apparel, accessories, and footwear for large companies from 19 July 2026, with medium-sized companies following from 2030. If your brand sells into the EU market, that deadline is now.
The conversation around fashion's environmental footprint tends to focus on production processes, shipping miles, and fiber choice. Those matter. But overstock is the environmental cost that sits entirely within a brand's operational control, and it begins with a forecasting problem, not a sustainability team problem.
Key Takeaways
- The EU bans destruction of unsold textiles for large companies from 19 July 2026, and for medium-sized companies from 2030
- Between 264,000 and 594,000 tonnes of textiles are incinerated or landfilled in the EU each year without ever being sold, generating roughly 5.6 million tonnes of CO2, comparable to Sweden's total net emissions in 2021 (Global Fashion Agenda; European Commission)
- Fashion accounts for approximately 10% of global carbon emissions, more than international flights and maritime shipping combined (McKinsey, Fashion on Climate)
- Overstock is primarily a demand planning failure: brands that reduce demand variability in their forecasts produce less excess inventory at source
- Reducing overstock at source is more environmentally effective than any downstream circularity measure: you cannot recycle your way out of overproduction
See how Metreecs helps fashion brands reduce excess inventory through demand planning.
The scale of fashion's overstock problem
Fashion is one of the most resource-intensive industries on earth. It accounts for approximately 10% of global annual carbon emissions, exceeding the combined emissions from all international flights and maritime shipping (McKinsey and Company, Fashion on Climate). It also consumes a significant share of global wastewater through dyeing and finishing processes.
Of the roughly 100 billion garments produced globally each year, an estimated 92 million tonnes end up as waste (Ellen MacArthur Foundation). At current trajectories, that figure is projected to keep climbing toward the mid-2030s.
Within that waste total, unsold inventory is a distinct and specifically avoidable category. European Commission figures put destroyed unsold textiles at 4-9% of everything placed on the EU market, or 264,000 to 594,000 tonnes incinerated or landfilled per year in Europe alone.
That is not waste from worn garments. That is waste from overproduction: units that were manufactured, shipped, warehoused, and then destroyed because demand was never there for them.
The EU regulation that changes the calculus
For brands with operations or sales in the EU, the environmental argument against overstock now has a compliance dimension.
The Ecodesign for Sustainable Products Regulation (ESPR) includes a ban on the destruction of unsold consumer goods, including textiles and apparel. The ban applies to large companies from 19 July 2026. Medium-sized companies face the same requirement from 2030. Small enterprises are currently exempt.
The regulation covers any product placed on sale in the EU market, regardless of where the brand is headquartered. A brand selling through an EU e-commerce channel, a manufacturer selling wholesale to EU retailers, or a fashion group with an EU subsidiary all fall under scope if they meet the size threshold.
The practical consequence is significant. Brands that have historically managed end-of-season overstock by incinerating or landfilling unsold units no longer have that option legally for EU-market goods. The alternatives available, including resale, redistribution, repair, recycling, and donation, all carry their own costs and operational complexity.
This makes overstock prevention, specifically reducing the volume of excess inventory before it accumulates, a more commercially rational strategy than overstock management after the fact.
Where overstock comes from: the forecasting root
The instinct is to frame overstock as a sustainability problem or a regulatory compliance problem. It is first a demand planning problem.
Fashion brands produce excess inventory for one primary reason: the pre-season forecast did not match actual demand closely enough, and production commitments were made on the basis of an inaccurate model.
The structural drivers of that gap are well understood.
Long lead times. Most fashion supply chains require production commitments four to six months before the selling season. A brand producing for autumn/winter must place fabric and manufacturing orders in spring, based on a demand forecast that will not be validated until autumn. Any shift in consumer behavior, competitor actions, or macroeconomic conditions between order date and sell date produces a gap.
Minimum order quantities. Suppliers enforce minimums that incentivize brands to round up. A model that needs 800 units ends up with 1,200 because the minimum order quantity is 1,000. The extra 400 units enter the system before the season starts.
Demand variability at the product level. Aggregate forecasts look reasonable. Product-level demand, particularly at the size-color level across a multi-store network, is much harder to predict. A style forecast to sell a set number of units can end up badly distributed relative to actual store-level demand.
Open-to-buy (OTB) pressure. Buying teams working within an OTB budget often allocate spend based on last season's performance. If a category overperformed last year, the OTB for that category increases. The assumption that last season predicts this season produces systematic overbuying in categories where trends shift.
Each of these drivers is addressable through better demand planning. None of them are addressed by improving end-of-life disposal options.
A planning failure with a visible cost
A pattern that shows up repeatedly in post-season buying audits: a multi-store fashion brand runs the same OTB model season after season (prior-year sales by store, adjusted by a flat growth factor, distributed across categories by a percentage split set years earlier). At the end of a season that ran warmer or colder than historical averages, the audit shows a meaningful share of the buy sitting in the warehouse as end-of-season stock. Some of it is close enough to sell at markdown; a smaller core, in sizes that never moved, has no realistic resale path.
Sustainability teams often start exploring donation and recycling options for that residual only once it exists. That conversation is valuable, but it happens after the problem has already been created.
The root cause is usually visible in the OTB model itself: a growth factor applied uniformly across categories regardless of how a season actually trended. The demand signal is often available within the first weeks of the selling season, but the production commitment was placed months earlier and cannot be undone.
Tightening the initial forecast, and reducing days inventory outstanding on slow-moving categories, prevents a meaningful share of that end-of-season volume from accumulating at all.
The three disposal routes and what they actually cost
When a fashion brand ends a season with unsold inventory, it faces three broad options: sell it at markdown or through resale channels, move it through redistribution or donation, or destroy it through landfill or incineration. The relative cost of each option is shifting.
Markdown sales convert overstock to revenue at reduced margin. A unit that cost 40 euros to produce and retails at 100 euros may clear at 50 euros in a sale, recovering 10 euros of margin instead of 60. The product has done its job, at the cost of brand positioning in that category.
Resale and circular channels are growing in availability but carry operational overhead: sorting, categorization, pricing, logistics, and platform fees. That overhead is meaningful, but it does not offset the cost of having produced the unit in the first place.
Destruction has historically been the lowest-friction option for truly unsaleable stock: units in colors or sizes with no buyer, heavily damaged returns, items held too long. The ESPR regulation eliminates this option for large EU-market brands as of 19 July 2026.
The math across all three routes points in the same direction: the most economically and environmentally efficient intervention happens before the excess unit is produced, not after it enters the disposal pipeline.
How demand planning reduces overstock at source
Preventing overstock is fundamentally a forecasting accuracy problem. The closer a brand's pre-season demand signal matches actual sell-through, the less inventory enters the system beyond genuine demand. This is the core of what demand planning systems are built to solve.
The mechanisms that improve forecast accuracy at scale are well established.
Store-product level forecasting. Aggregate category forecasts mask the distribution of demand across stores and sizes. A forecast that resolves to the store-product-size level produces far more accurate OTB inputs than one that operates at category or brand level. The sell-through rate signal at this granularity shows where to allocate stock and, equally important, where not to.
In-season demand sensing. The first two to three weeks of a fashion season contain the strongest signal about full-season sell-through. Brands that feed early sell-through signal back into their planning models, adjusting OTB and transfer decisions before the markdown window, structurally reduce end-of-season excess.
Safety stock optimization. Safety stock held against demand variability should reflect actual variability at the product level, not a flat percentage applied across the buy. A style with low demand variability and high sell-through consistency needs less safety stock than a trend-driven item with uncertain demand. Calibrating safety stock to observed forecast error at the product level reduces unnecessary buffer inventory across the range.
Reorder point discipline. Setting reorder points that reflect actual lead times and service level targets prevents both stockouts and the overcorrection that creates excess. Many fashion brands hold more inventory than their service level targets require because reorder point calculations use average lead times rather than actual lead time distribution.
Brands that adopt these mechanisms tend to see a meaningful reduction in total inventory within the first few seasons, driven by tighter OTB planning, better allocation, and fewer end-of-season overstock positions.
When the sustainability case and the financial case align
A pattern that recurs across multi-brand fashion groups launching sustainability initiatives: the first wave of measures focuses on downstream fixes (packaging reduction, recycled fiber sourcing, a resale pilot with an external partner), while the demand planning side of the business is treated as a separate workstream.
In practice, the two are the same problem viewed from different angles. A brand's own sell-through data usually shows that a meaningful share of the buy remains unsold at end of season: most of it clears at markdown, some gets redistributed, and a smaller residual has no commercial path. Post-regulation, that residual needs an alternative route at real operational cost.
The internal case for improving forecast accuracy is straightforward once framed this way: eliminating a portion of the buy that would otherwise become end-of-season excess produces cost savings that typically outweigh the operational cost of downstream sustainability measures alone. The strongest sustainability business cases now combine both, with the demand planning investment approved on financial grounds first and environmental grounds second.
The logic is direct: a unit that is never overproduced has zero end-of-life environmental cost. A unit that reaches landfill carries its full carbon cost of production, transport, and warehousing, with the disposal impact added on top.
The planning vocabulary of overstock prevention
Days inventory outstanding (DIO) measures how long inventory sits before converting to revenue. A DIO that increases season over season is an early indicator that overstock is accumulating faster than the business is clearing it.
Product-level sell-through rate is the primary signal for identifying which units are at risk of becoming overstock. A sell-through rate well below half at the midpoint of a standard selling window is a strong indicator that the unit will not sell through at full price.
Demand variability is the spread of actual demand outcomes around the forecast. High demand variability at the product-size level is the core driver of both stockouts and overstock: the same forecast error that leaves a brand short on one size leaves it long on another.
Forecast accuracy is the standard measure of planning quality at scale in fashion, commonly tracked through WMAPE (weighted mean absolute percentage error). Reducing forecast error by a meaningful margin removes a proportional share of misallocated units from the buy, which is the most direct route to reducing overstock without sacrificing service level.
Open-to-buy (OTB) is the buying budget remaining for a planning period. OTB set against inflated demand forecasts is the most direct mechanism for generating overstock: the budget authorizes a buy that genuine demand will not absorb.
Safety stock is the buffer held against demand variability and supply lead time uncertainty. Safety stock calibrated to actual product-level variability reduces excess without increasing stockout risk.
FAQ
Does reducing overstock make a meaningful environmental difference?
Yes. The fashion industry destroys between 264,000 and 594,000 tonnes of unsold textiles in the EU alone each year, generating roughly 5.6 million tonnes of CO2. Each of those units carried the full carbon, water, and chemical cost of production before being destroyed. Reducing the volume of excess inventory produced reduces those upstream costs at source, which is more impactful than any downstream disposal improvement.
What does the EU ESPR regulation require of fashion brands?
The Ecodesign for Sustainable Products Regulation bans the destruction of unsold consumer goods including apparel, clothing accessories, and footwear for large companies selling in the EU market from 19 July 2026, and for medium-sized companies from 2030. The ban applies regardless of where the brand is headquartered. Brands must find alternative routes for unsold stock: resale, repair, recycling, redistribution, or donation.
Is AI demand planning a realistic solution for mid-sized fashion brands?
Yes. Mid-sized brands managing hundreds to thousands of active products across dozens of stores can access systems that provide store-product level forecasting and in-season demand sensing, typically recovering the cost within a couple of seasons through reduced markdown and lower overstock volumes.
What is the connection between overstock and carbon emissions?
Every overstock unit carries the embedded carbon of its production: fiber cultivation or synthesis, spinning, dyeing, cut and sew, finishing, packaging, and transport. When that unit is destroyed without being worn, its entire carbon cost is wasted. The fashion industry is responsible for approximately 10% of global annual carbon emissions, and a meaningful portion of that footprint is attributable to overproduction and eventual destruction of unsold goods.
How does better demand planning interact with circular economy initiatives?
They are complementary but not equivalent. Circular economy measures, including resale, repair, rental, and recycling, reduce the environmental cost of units that do enter the system as excess. Demand planning reduces the volume of units that become excess in the first place. The most effective approach combines both: a tighter buy reduces the volume requiring circular interventions, and better circular infrastructure handles the residual overstock that remains.
Can overstock prevention replace sustainability certifications and green sourcing?
No. Fiber choice, manufacturing processes, chemical management, and worker welfare are independent sustainability dimensions. But overstock prevention is the lever with the most direct financial return and the clearest operational mechanism for a planning team. It does not require changing suppliers or reformulating dyeing processes. It requires better demand planning, tighter OTB discipline, and more accurate product-level forecasting.
Conclusion
The environmental cost of overstock in fashion is not abstract. It is measured in the 264,000 to 594,000 tonnes of unsold textiles destroyed in the EU each year before a single customer ever wore them. It is embedded in the carbon footprint of every unit manufactured against a forecast that was too optimistic and a minimum order quantity that rounded up.
The EU's ESPR regulation, effective for large companies from 19 July 2026, removes the lowest-friction disposal route for that excess. What remains is prevention.
Prevention starts in the planning cycle, with OTB decisions grounded in product-level demand signals rather than prior-year growth factors. It continues in-season, with sell-through monitoring that catches imbalances before they harden into end-of-season overstock. It scales through AI-driven inventory optimization that processes the store-product-size combinations that manual planning cannot handle.
If your brand's end-of-season analysis regularly shows units with no commercial path, the answer is not better waste routing. Book a demo to see how Metreecs reduces the volume of inventory that enters the problem in the first place.
Sources
Our articles
Own your operations with AI Intelligence
.svg)