Luxury: fund the advisor's relational time before rolling out client-facing AI
Luxury clients, especially top spenders, already use AI before buying; brands struggle to industrialize it. Executives must decide where AI frees high-value advisory time.
Key takeaways
- 82% of top-spending clients used AI at their last purchase versus 28% in the lowest-spending segment (Bain × Colbert).
- 22% of houses rank AI in their top three priorities (5% in 2024), yet fewer than 20% of executives see significant impact.
- Only 9% of houses have scaled AI support for advisors with measurable impact.
- The right indicator is revenue per advisory hour on high-potential clients, not usage rate.
The 2026 Luxury and Technology report, published on 30 June 2026 by Bain & Company and Comité Colbert, describes a reversal: AI has become a strategic priority for luxury, but clients are moving faster than brands.
Priority figures: 22% of houses and groups rank AI in their top three priorities for the next three years, versus 5% in 2024; 61% place it in their top ten (50% in 2024). Deployment grows by function (support 6% → 31%, operations 10% → 19%, customer-facing 16% → 21%), yet fewer than 20% of executives see significant impact.
Client figures: average AI usage reaches 64% in China, 54% in the United States and 27% in France. Above all, 82% of top-spending clients used AI at their last purchase versus 28% in the lowest-spending segment; 47% of in-store buyers used it before visiting, and 97% plan to use it again.
Economic mechanism: information, once scarce and held by the advisor, becomes abundant and free. The advisor's value shifts to interpretation, trust and personalization. This relational work is hard to automate: following Baumol's logic, the relative cost of labor-intensive services tends to rise. AI should therefore be framed not as a substitute but as a way to make each advisory hour more productive.
The gap shows on the brand side: only 9% of houses have scaled AI assistance for advisors with measurable impact, and 22% are getting first results. Bain also notes that transformation cannot simply be declared: it requires vision, governance, talent and data infrastructure.
Implications by type of player. House with its own network: the advisor serves a better-informed client, so visit preparation and client memory are the first source of value. House with mixed distribution: the data gap between own boutiques and retailers limits the realistic pilot scope. Cosmetics: same logic, with shorter and more frequent advice, hence a different trade-off between advisor copilot and client assistant.
Limits. The release gives neither sample size nor survey method, which sit in the full report. A client's AI use says nothing about their propensity to spend more. And no public data yet quantifies the revenue gain of an advisor copilot; it has to be measured internally.
Bottom line for executives: the question is no longer whether to put AI in the boutique, but where it frees advisory time on the highest-value clients, and which indicator decides the rollout.
Our recommendations
- 1
Measure the first 10 minutes
Retail director, within 30 days: observe in 2 or 3 boutiques what the advisor does with an already-informed client. Identify the tasks the client has already done alone.
- 2
Pilot a copilot in one boutique
Client director and CIO, within a quarter: one visit-preparation copilot, one high-potential client scope, one control group.
- 3
Set a single indicator
General management: revenue per advisory hour on the target segment. Without it, usage rate hides the absence of impact.
- 4
Lock down client data
DPO and CRM head: consent, client-record quality and usage rules before any rollout; a copilot is only as good as client memory.
Analysis by
Cardan-AI Intelligence
Our research and analysis unit, dedicated to applied AI for business, industry and regulatory compliance.
Let's talk about your next competitive edge
Thirty minutes to identify the two or three use cases in your operations that pay for themselves within the first year.