Mytheresa has scored its future VIPs since 2017. Your boutique can too

In June, Glossy's Luxury Briefing sat down with Michael Kliger, chief executive of LuxExperience, the parent of Mytheresa, Net-a-Porter, Mr Porter and Yoox. The whole interview is worth reading, but one passage deserves to be pinned above the desk of every boutique owner. Asked how the group uses AI, Kliger described predictive models Mytheresa has run since 2017 to identify which first-time shoppers will become its most valuable clients, long before their spend says so.

The playbook

The signals Kliger listed are worth quoting: a customer's first purchase, their search behaviour, payment method, products and brands purchased, shipping and billing address, even the time of day a transaction is made. The product itself, he said, remains one of the clearest tells. An expensive dress suggests a client building a wardrobe; an expensive handbag may be a one-off.

When the model flags a first-time shopper as high future value, Mytheresa moves early: preferred service, complimentary shipping and returns, sometimes a personal shopper, before the client has spent much at all. “We start to act as a company as if she or he is already a good customer,” Kliger said.

That is the hidden VIC, described by the largest multi-brand luxury retailer in the world. The client whose till receipts look ordinary and whose signals say otherwise, recognised and treated accordingly, before a competitor gets there.

Nine years of compounding

Mytheresa has run this discipline since 2017, and it shows in the numbers the group reports. In the quarter announced this May, Mytheresa remained LuxExperience's strongest business, with net sales up 9.9 per cent at constant currency and up 33.8 per cent in the United States, in a luxury market that has been anything but easy. Reading future value early, and acting on it, compounds.

Kliger's second insight is about time. Older recommendation models, he noted, lean on what a client did a year ago; his team now weights the last three weeks. “I'm less interested in what you did a year ago. I'm more interested in what you did in the last three weeks,” he said. A client's signals are freshest right after they act, which is why the follow-up that lands the same week beats the campaign that lands next season.

Editors and associates

For all the modelling, Kliger was emphatic about where the machine stops. Mytheresa's buying is led by editors, its events are curated by people who know that two clients share a background in biotech and should be seated together, and its ambition for AI in clienteling is to take the desk work away, “from pulling customer information to preparing recommendations, so personal shoppers can spend more time directly with clients. We still believe the human touch makes a huge difference at that level.”

We could not have put Halia's own design brief better. The engine finds the client; a person at your brand makes the call and builds the relationship.

What this means for a boutique

Mytheresa built this capability in-house, with a data team, over nine years. That is the honest cost of the playbook, and it is why, until now, it has belonged to retailers with nine-figure revenues.

Halia is that capability for a single boutique. It reads the same kind of signals in your own customer data, grades every client, and surfaces the ones whose future value their spend has not yet revealed, so your team can act as if they are already a good customer, because they are about to be. Your data stays in your store, your associates get the time back, and the playbook the biggest house in luxury e-commerce has validated for nine years starts working for yours.

Quotes and figures are from Glossy's Luxury Briefing of 12 June 2026, reported by Zofia Zwieglinska. The reading, and any opinions, are our own.