- A leading health insurance provider in the US was looking to understand the effectiveness of their marketing activities across products, sales channels and DMAs, for a highly seasonal product.
- They also wanted to understand the difference in effectiveness of allocating spend to their own brand versus partner marketing.
- Analysis of marketing’s long-term impact on DMA level customer enrollments and LTV would further enhance optimization of spend in brand building and performance programs.
- Advanced dynamic multi-level hierarchical bayes models were built to understand the dynamics between the client's various products, sales channels and types of marketing spend (direct vs. partner), within and across DMAs. This enabled them to identify what portion of sales were driven by advertising vs. product enhancements in the short-term. Dynamic brand models were then developed to identify the impact of brand equity and customer perceptions in the long-term.
- Leveraging the modeling results and the Marketscience SimOpt tool, we optimized the client’s marketing mix across online and offline channels and DMAs. We also provided them with detailed insights around specific media performance improvement strategies balancing the short- and long-term.
- The insights suggested some significant changes to the balance of own versus partner spend so we leveraged the dynamic cross-market structure of the models to support a test and learn approach to implementing the changes and further refine the optimization
- +5% increase in enrollments driven by measuring long-term marketing impact versus looking at short-term only
- +12% ROI increase when reallocating same budget towards brand messaging and related channels