Reduce time to measure ad effectiveness by more than 80%
The Business Challenge
- The client was spending over $1 billion annually on mass media and digital marketing globally without successfully measuring ROI.
- Fractal needed to accurately measure the impact of daily data on sales in the online and offline channels, quantify the effect on 10 channels influencing the different moments of truth in the shoppers’ journey, and create a standardized cross-media optimization approach and scale it across markets.
- Developed over 3 million models in less than 2 months to finalise a highly robust ensemble model
- Modelled complex (some non linear) relations using 5 different techniques – Multivariate adaptive regression splines, Generalized additive models, N-Parameter logistic regression, Polynomial functions and Power curves
- Developed Bayesian Belief Network based approach to model cross media effects and understand complex network relationships between marketing channels.
- Automated the entire modelling process to enable standardization across business lines and regions as well as to operationalize at scale
We developed an integrated 3 step cross media optimisation framework*
The models are ‘live’ in select regions currently and are being implemented in other geographies. The results beat the expectation in terms of increase in traffic (~1%), revenue impact (~1.1%) and reduction in marketing budget (0.65%)
Identifying decay and saturation levels of TV advertising and suggested optimal execution leading to:
- Capturing indirect effect of TV and display ads through SEO and partner networks.
- Increasing TV spending and flighting strategies from current to suggested optimal levels
- Reduced overall time to measure ATL marketing effectiveness by more than 80%
- Enabled standardization of Marketing ROI measurement across business lines and regions
Our approach (and solution) was identified as the best amongst 10 competitive solutions that the client evaluated for their business.