Case Study
Session-Level Pricing
+34% revenue for a Swiss tourism marketplace.
Client
Swiss tourism e-commerce platform
Market
Swiss travel & commerce
Scale
Ski, hotel, activity and restaurant bookings
Duration
12 months
Team
Senior-led ML & platform team
Engagement model
Embedded Partnership
Delivery hub
Zagreb
Stack
Custom ML, real-time pricing engine, LLM content generation, A/B framework
Capability
Platform & Managed Services
Challenges
Challenges
Static pricing
Fixed discounts failed to optimise revenue or convert price-sensitive users.
Diverse demand
One rule served luxury travellers and budget families equally badly.
Competitive pressure
Aggressive pricing from emerging booking platforms.
The solution
Real-time behavioural analysis
Live customer profiles built from interactions during the session.
Dynamic price adjustment
Prices and offers adjusted mid-session to maximise conversion within margin limits.
Personalised content
LLM-generated offer copy and voucher descriptions tailored to the customer.
Cross-product intelligence
Behaviour analysed across categories to lift average order value.
The honest section
What we would tell you before starting
Session-level pricing only works with enough live traffic to learn from; a low-volume catalogue would not have produced signal worth acting on. We validated that the platform’s volume supported the approach before building the pricing engine.