Sign up

Miroglio Fashion – Case Study

Miroglio Fashion was struggling. Uneven product sales across regions and stores, excessively long product shelf life, significant promotional pressure from market trends and competition.

Could advanced analytics help reverse the trend? Miroglio implemented a new dynamic exchange of inventory across stores and warehouses, using the Evo Transfer product. The result: increased margin and accelerated sell-through.

Better coverage of granular demand and greater efficiency in the supply chain resulted from the collaboration between store managers and a new generation of machine systems based on Prescriptive Artificial Intelligence (AI).

Increased revenues +16.2% by introducing
Evo’s inventory exchange

How Miroglio Fashion

+16.2%

94%

+€1M

Revenue
Growth

Allocation
Efficiency

Incremental
Margin

Context: the history of Miroglio Fashion

Miroglio Fashion is Italy’s third-largest retailer of women’s apparel, with €520m+ annual revenues from thousands of stores across 11 brands. After building its first factory in 1947, it grew consistently through the 2000s.

By the 2010s, however, revenues were falling. While some stores continued to thrive, performance diverged between top and bottom-earning stores.

This uneven performance exacerbated existing supply chain issues, increasing both stockouts and overstocks.

The challenge: uneven demand patterns

Francesco Cavarero, the Chief Information Officer of Miroglio Fashion, was searching for technology solutions to help reverse these downward trends. He thought analytics could be effective, but remained hesitant.

I wasn’t sure that we could build a useful analytical model that worked within our position in the industry. Anticipating sales has always been challenging in fashion. Trends and customer needs change all the time, making demand volatile.

At Miroglio, short-lived items sometimes only sell a few units per week. I was concerned that forecasts based on so little data may not be dependable or deliver the desired ROI.

-Francesco Cavarero, Miroglio CIO

Additional, critical problems stood in his way:

  1. Slow inventory turnover: Miroglio’s relative size and industry position naturally results in some slow-sellers that reduce the usable floor space in stores.
  2. Traditional push approach to replenishment: Real sales could differ from forecasts by as much as 40–50%. Therefore, the traditional replenishment model was based on push driven by past sales: “sell-one, send one”.
  3. Internal resistance: Cavarero knew that some managers at Miroglio would fear losing autonomy and control, trusting their own intuition over the perceived black box of analytics.

Moreover, Cavarero wanted to retain the valuable insights that came from the experience of store managers who had an average tenure of 15 years and knew their customers well.

The solution: Evo’s inventory exchange, based on a new human-machine alliance

Miroglio partnered with Evo in 2016 to implement Evo Transfer, the responsive supply chain solution. Prescriptive Artificial Intelligence generated rapid-response recommendations to market conditions, targeting impact instead of accuracy.

We had little time to waste in returning Miroglio to an upward growth trajectory. It would take us months, if not years, to build a team of qualified data scientists in-house.

Evo had both the technical skills and industry experience needed to deliver ROI quickly, so we decided to partner with them to develop innovative solutions for our core planning and distribution processes.

Evo implemented a dynamic inventory management system: direct inventory exchange, based on a granular analysis of demand at the product and size level to rebalance the performance across stores.

This approach relied on a pragmatic, daily human-machine alliance:

1. Track company and market data

Evo combined historical sales data with extensive market, weather and social media data to improve performance: for example, monitoring 121,000 competitor products prices and attributes and the consumer behaviour of 18% of the Italian population from all the pilot stores.

2. Integrate the store manager input

After calculating demand for each item and store, Evo Transfer recommends shipments. All store managers can then use their local knowledge to make changes and swap products or sizes within a given window of time without changing the total shipment volume. Evo Transfer automatically adjusts shipments accordingly based on systemwide availability.

3. Automate the weekly inventory transfers

Evo Transfer provides inventory lists for warehouse shipment and direct cross-shipment between stores — all to minimize logistical effort and costs while increasing margin and sell-through.

To measure impact, Evo Transfer was initially deployed in 35 stores of two brands during a rigorous one-month A/B pilot test.

We needed proof positive from the earliest stages of the project that a new approach would increase margin without overly complicating store operations. Anything less, and we’d never secure internal buy-in.

Pilot impact: +16.2% revenue growth

The initial pilot generated margin of €300,000 within the limited scope of the test, and ultimately delivered +16.2% revenue growth.

The initial results already delivered above and beyond our hopes for the initiative. Evo Transfer started to turn our KPIs around and show returns from week 1.

Store managers were critical to success. Their experience allowed them to predict popular items before the algorithm picked up the pattern. They saw customer enthusiasm first-hand, generating vital new data ahead of the curve.

Analytics bear out the success of store manager involvement: +5p.p. sales forecast accuracy with store manager input for an overall 94% allocation efficiency, a +25p.p. improvement over the original system.

Our store managers felt excited about the new replenishment process. On average, 72% of our store managers provided input throughout the pilot, and they were pleased to see their requests honoured by the system.

Our local staff have long been a vital part of our success. Their engagement let each store manager multiply their individual impact.

Long-term results: €1 million/month incremental margin

After the successful pilot, Cavarero rolled Evo Transfer out across all brands and regions: during its 5+ years in operation, it has already managed millions of pieces of inventory.

Post-roll-out, Miroglio achieved €1 million/month incremental margin with a reduction in leftover inventory and 23% fewer stockouts.

Store managers continue to enthusiastically participate, with over 80% of managers editing their proposals every week and positive feedback.

It has been a winning partnership

Francesco Cavarero

Evo has been our top innovation of 2016. The impact far exceeded our expectations, delivered professionally.
The longer we let this replenishment system learn, the more our impact grows.

Want to learn more?



Hey! Was this page helpful?

We’re always looking to make our docs better, please let us know if you have any
suggestions or advice about what’s working and what’s not!

Send Feedback