AskBosco · Marketing mix modelling · 2025
Marketingmixmodelfor Sigma Sports
Found £80k+ of extra revenue by moving budget between channels, with no extra spend.
£80k+Incremental revenue identified
01 / 03
Every split of the budget earns something
Each point on the surface is one way to divide the same budget between channels. Its height is the revenue the model expects.
02 / 03
Returns flatten as a channel saturates
Read from above, each contour joins splits that earn the same. The best plan is a balance between channels, not a bet on one.
03 / 03
Move the budget, not the total
The optimiser climbs to the best split. For Sigma Sports, that meant £80k+ of potential revenue with the same spend.
The project
- Client
- Sigma Sports
- Role
- Data Scientist, AskBosco
- When
- 2025
- Discipline
- Marketing mix modelling
- Tools
- Python
- Vertex AI
- GCP
- XGBoost
- Time series
The situation
Sigma Sports needed to predict campaign ROI and revenue by channel, and to know where future budget should go.
What I built
- A marketing mix model that predicts campaign ROI and revenue by channel
- Dynamic channel attribution
- Budget optimisation logic for future spend
- Automated retraining pipelines
What changed
The optimised budget allocation identified £80,000+ of potential incremental revenue in Q4 2025, without additional marketing spend.
- £80k+Incremental revenue identified
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