Share of real cases from customers
The Client: Luce di Milano – a prestigious department store with 9 locations across Milan, Rome, and Florence. They have been selling luxury home fragrances for over 20 years, but their candle department was bleeding margin. Every season, they ordered based on “last year’s numbers plus 10%” – a crude method that left them with 38% unsold inventory after each holiday.
The Pain Point: Their previous supplier (a French manufacturer) required orders 120 days in advance, with no flexibility to adjust quantities or scents after production started. Luce di Milano’s buying team had to guess which seasonal scents would sell – and they guessed wrong repeatedly. In 2024, they over‑ordered 5,200 units of “Chestnut & Clove” and under‑ordered “Amber & Saffron” (which sold out in 3 weeks). The unsold stock cost them €78,000 in write‑offs and warehouse fees. Their buyer, Matteo, told us: “I need a supplier that helps me predict demand, not just take my order.”
Our Solution: We didn’t just produce candles – we provided a predictive analytics service built on our historical sales database (1.2 million candles sold across 40+ retailers in similar climates).
Scent Performance Index: We analysed 3 years of sell‑through data for Italian retailers and created a “seasonal heatmap”. For the winter season, we ranked 15 scents by popularity in Milan vs. Rome vs. Florence – because regional preferences differ.
Flexible Order Windows: Instead of one 120‑day order, we offered 3 “forecast‑to‑order” checkpoints: at 90 days (commit to 60% volume), at 60 days (add 25%), at 30 days (final 15%). Between checkpoints, Matteo could swap scents based on early pre‑orders.
Buffer Stock Program: We held an additional 2,000 units of his top 3 scents in our Rotterdam hub – available for emergency top‑ups within 7 days.
The Process:
Day 1 (August 10): Matteo sent his initial wishlist – 18,000 candles across 6 scents for the Christmas season. Our data team ran a comparison: “Chestnut & Clove” historically underperformed in Florence, while “Vanilla & Cinnamon” overperformed in Milan. We recommended reducing Chestnut by 40% and boosting Vanilla by 30%.
Day 10: Matteo agreed to our adjusted mix: 16,000 candles – 5 scents, with 60% committed.
Day 60 (September 20): At the second checkpoint, Matteo saw that online pre‑orders for “Spiced Orange” were 2.5x higher than expected. He swapped 1,200 units of “Pine” to “Spiced Orange” – we accepted the change without penalty.
Day 90 (October 25): Final checkpoint – he added 1,500 emergency units of “Amber & Saffron” from our buffer stock.
Delivery: First batch arrived October 15, second November 5, final November 25 – perfectly staggered.
The Result:
Unsold seasonal stock dropped from 38% to 8.2% – saving €67,000 in write‑offs.
Matteo’s candle revenue grew by 22% because they had the right scents in the right stores at the right time.
He placed a spring order with us 45 days earlier than usual – trusting our data model.
His bonus: he now uses our seasonal heatmap as an internal planning tool for other categories.
Why Our Factory Stands Out:
Data‑driven consulting – we don’t just take orders; we help you place the right order.
Multi‑checkpoint flexibility – no penalties for changing scents up to 30 days before production.
Regional preference database – we track buying patterns across 22 European cities.
For department stores like Luce di Milano, we turn guesswork into science – and unsold stock into profit.
