An AI-powered travel recommendations engine
A machine learning recommendation engine that lifted booking conversion by 45%.
- Client
- Wanderlust AI
- Industry
- AI & Data
- Location
- San Francisco, USA
- Timeline
- 5 months
- Team
- 5 people
- Conversion increase
- 45%
- ROI on recommendations
- 3.2x
- Recommendations a day
- 12M+
- Higher order value
- 22%
The challenge
- New users have no history to personalise from
- Limited training data at the start
- Personalisation had to run in under 100ms
Our approach
- A hybrid model combining collaborative and content-based filtering
- A real-time feature store for behaviour signals
- Transfer learning to handle cold starts
The result
- Booking conversion up 45%
- Average order value up 22%
- 12M+ recommendations a day under 100ms
- Services delivered
- Machine learningData engineeringBackend development
- Technology
- PythonTensorFlowLaravelRedisAWS SageMaker
Keep reading
More case studies
Onboarding Pass
From Today's setup to Your next case study
Have a similar challenge?
- Passenger
- Your team
- Gate
- First call
- Seat
- 1A
- Boarding
- Now
Tell us what you're running today and where it hurts.