FRAME · Computer vision · 2024
AIimagescoringfor cottages.com
Scored listing photos to lift bookings: +4% on the weakest 20% of listings, at half the production cost.
+4%Bookings on the weakest 20% of listings
01 / 03
Every listing, scored
Each tile is a listing. The model scores its photos for quality.
02 / 03
Find the weakest 20%
The lowest-scoring listings get specific suggestions for better photos.
03 / 03
Lift them
Those listings gained 4% in bookings, and the system cost half as much to run.
The project
- Client
- cottages.com
- Role
- Data Scientist and AI Engineer, FRAME
- When
- 2024
- Discipline
- Computer vision
- Tools
- Computer vision
- Python
- Web app
- Image analysis
The situation
cottages.com wanted higher conversion rates on its listings, starting with the quality of their photos.
What I built
- Led the technical side of the project
- Image-based quality metrics and scoring algorithms
- A web app that gives each listing actionable suggestions
What changed
Added 4% booking uplift on the worst 20% of listings, and cut the AI system's production costs by 50%.
- +4%Bookings on the weakest 20% of listings
- 50%Lower production cost
Hiring for marketing measurement or customer ML in the UK?
Eligible to work in the UK, no sponsorship needed. Based in London.
The CV is a two-page PDF, rebuilt from this site whenever it changes.
Jihar Gifari · London · jigifari@gmail.com · +44 7985 687426