GermanyRetail and e-commerce2017 – ongoingLuxExperience (Mytheresa)
Predictive models that spot future high-value shoppers
Mytheresa uses predictive models to flag first-time buyers likely to become VIPs
Problem
Cost-per-click figures say little about whether paid marketing brings in customers who will spend a lot over time. Personal service for top clients was also slow to reach new shoppers.
Approach
LuxExperience, the parent of Mytheresa, has used predictive models since 2017. They score a first-time shopper on the first purchase, search behaviour, payment method, brands bought, addresses and time of day. If a shopper looks likely to be high-value, Mytheresa may offer preferred service, free shipping and returns, or personal shopping early. Its recommendation models now weight the last few weeks of activity more than older history.
Results
No published figures
Takeaway
Scoring customers on future value let the retailer judge marketing spend and service offers by long-term worth, not by clicks.
Categories
- Size
- Large enterprise
- Industry
- Retail and e-commerce
- Business function
- Marketing and sales
- Innovation type
- Process innovation
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Glossy describes how LuxExperience uses predictive models, in place since 2017, to estimate the value of first-time Mytheresa shoppers and trigger early VIP treatment. It also covers recommendation models that favour recent activity. The article gives no results for the AI itself.
Full details are in the original sources.
