One of the top luxury men’s fashion retailers in Europe achieved +4% like-for-like revenue growth while reducing overall stock by 12% – after just seven weeks of applying Evo’s replenishment system.
High-end retailers often carry excess stock, as even missing one sale carries a significant opportunity cost. However, the more accurate automated demand forecast powered by Evo allocated the right products to the right stores at the right time, reducing wasteful overstock and freeing up inventory for faster-selling stores.
The problem of supply chain complexity—and excess stock
Carrying the wrong stock in the wrong stores at the wrong time is incredibly costly, especially in the high-margin luxury market. Marking down unsold merchandise hurts the bottom line. Moreover, high levels of stocks across stores reduce retailers’ flexibility to adjust to customer demand throughout the season. Since carefully designed luxury boutiques can only carry so much merchandise, poor sellers take valuable space away from better options.
A leading EU luxury men’s fashion retailer needed to reduce complexity in their supply chain. They already had over 100 retail locations in their biggest market, Italy, and were rapidly growing internationally. Basic replenishment systems had long been in place, but they could no longer keep pace, as they were built during an earlier, slower period of the company’s growth.
Like many retailers, this menswear brand was consistently hamstrung by the limitations of its customer demand forecast. The company had no way of accurately estimating how much of each item would sell locally since customer segments and preferences varied widely from store to store. Stores would compensate for this variance by merely overstocking a wide variety of articles in order to chase even relatively low levels of sales.
Such levels of inventory are challenging for individual stores to manage: their nature as curated showrooms required displaying only a limited number of pieces at a time. While overstocking may be an effective way not to miss sales, it is inefficient. Low sell-through rates  of some articles in certain stores regularly led to unplanned overstocks. Better-performing stores, on the other hand, experienced undesirably high levels of fragmentation. Overall, the network was less able to meet customer demand. Business leaders struggled to identify patterns that would explain these inefficiencies. Inefficient allocation seemed inescapable.
The retailer decided to look for a partner to achieve a more granular forecast accuracy as a way to reduce inventory inefficiencies. They ultimately chose to work with Evo, an expert in AI-based supply chain automation.
We needed a more efficient and more accurate system that could deliver positive ROI fast. Evo guaranteed significant results in just a few weeks.
Evo distinguished itself by accurately forecasting daily sales at individual stores right down to the product and size. Evo showed that this more accurate store-level demand forecast would lead to greater diversification among stores, products, and customer segments relative to the past, in turn unlocking greater inventory efficiency. Essentially, their strategy was to reduce inventory as a means of increasing sales and revenue across the network.
Why reduce inventory?
It may seem intuitive that reducing stock in stores automatically increases the efficiency of inventory allocation. In reality, less stock often has the opposite effect. Lower stock allocation magnifies the impact of errors in the demand forecast.
The company had long struggled to prevent fragmentation. Some stores would run out of critical high-demand sizes, leading to missing sales, even despite relatively high inventory levels. As such, the promise to reduce stock did not initially appear a winning strategy, given its higher perceived level of risk.
How Evo’s revolutionary algorithm increases forecast accuracy at a granular level
Evo’s demand forecasts would have to be significantly more accurate to reduce errors and deliver results. Evo’s references looked solid enough to give the algorithm a chance, especially since those references showed a 72% average reduction in forecast errors. Thanks to a massive set of real-time data from a wide range of internal and external sources, trends and patterns could be identified earlier and more accurately, using the AI to do the heavy lifting.
Evo first integrated all the company’s product, store, and customer attributes into the system by matching the company’s internal data to its proprietary scalable data model. The Evo Replenish tool combined that enhanced data with Evo’s external data sources: market benchmarks, industry statistics, social trends and customer profiles — even the weather.
This much data, impossible to manage without dedicated tools, led to daily demand forecast at the granular level, down to the store, the product, and even the article size. With a more accurate customer demand forecast, overloading each store with stock proved unnecessary.
When companies know customers at the data level, it is almost possible to anticipate customer needs before they emerge. With this customer relevance, we can get the right products to the right stores at the right moment.
Evo did more than just provide a demand forecast. They also automated the logistics necessary to put the demand forecast into action. Based on what products it expected customers to seek, Evo Replenish mapped out stock transfers to distribute higher-demand articles to the right stores in the right sizes.
In addition to recommending the best warehouse-to-store replenishment flows, the algorithm also triggered store-to-store inventory transfers. Items not selling well in one location could be transferred to another store where the demand forecast expected higher sales: this stock would become profitable instead of dead inventory. The technology only allowed stock movements that generated profits accounting for the extra logistical costs, thus growing overall ROI.
The results: Reducing inventory increases sales
The Evo strategy quickly delivered results for the company. Within the initial seven-week A/B test, the automated replenishment system significantly outperformed the company’s original system.
The algorithm progressively reduced store inventory; by the final week of the test, stock levels at the stores using Evo Replenish were over 12% lower than at stores continuing to use the old system. Despite these changes, Replenish proved incredibly adept at anticipating and serving customers’ needs.
Revenues at stores using Evo Replenish increased by an average of 4%, and full-price sell-through increased by an average of 4 percentage points. The sell-through rates of the test stores outperformed the control stores at consistently increasing rates as stocks decreased. In other words, the longer Evo Replenish was controlling replenishment, the more it exceeded the old system’s performance.
One thing was clear to the retailer: the test was a major success. The impact on margin and overall revenues was significant— and the company moved the management of all inventory to the Evo Replenish system.
Sometimes less is more – really
For this luxury fashion brand, less really ended up being more. Within just seven weeks, Evo Replenish simplified inventory management, reduced overstock and fragmentation, and eased warehousing and storage issues — all while increasing full-price sell-through and revenue.
It has been exciting to see the initial impact on our margins. As we continue to work together, we look forward to even greater returns using Evo’s tools.
This luxury menswear company was eager to build on these first successes – and extend Evo’s AI across the rest of the store network and more.
 Sell-through measures sales divided by all available stock at the store, region, company level. When articles perform according to expectations, sell-through reaches 100% towards the end of the season.
About the author
Kaitlin Goodrich is Evo’s main storyteller who helps communicate Evo’s message to the world.
Kaitlin received her BS in International Affairs and Modern Languages at Georgia Tech and then an LLM in International Trade Law from the University of Turin. She worked in Latin America doing education outreach for U.S. binational centers and has since worked as a content writer for international clients.
In her free time, she likes to travel or curl up with a good book.