Situation
A leading pharmaceutical manufacturer sought to optimize promotional spending across channels and geographies for their NOAC (Novel Oral Anticoagulants) and DME (Diabetic Macular Edema) markets. The objective was clear: maximize profits by refining strategy and identifying high-impact promotional channels.
Challenge
The client was grappling with a mess of promotional data scattered across various geographies, making any meaningful analysis as elusive as a good hair day in a hurricane. They were clueless about the ROI from personal versus non-personal promotions, which is like trying to find your car keys in a haystack. Plus, they desperately needed scalable tools to give regional teams the power to execute strategies independently, without turning it into a bureaucratic nightmare.
Approach
Optimized Data Transformation was achieved through the application of advanced techniques, such as Linear-Log, Log-Log, and Interaction Variables, which significantly enhanced the readiness of data for analysis, ensuring that the insights derived were both actionable and precise.
- Enhanced Outcome Analysis: By employing sophisticated modeling techniques like Multivariate Additive Regression Splines (MARS), Ordinary Least Squares (OLS), and Partial Least Squares (PLS), we comprehensively evaluated the effectiveness of promotional strategies.
- Enabling Global Scalability: We developed reusable components within Jupyter notebooks to empower local teams, thereby facilitating the independent execution of Marketing Mix Modeling (MMx) projects across 30 diverse markets.
- Actionable Insights for Strategy: Our analysis revealed a superior Return on Investment (ROI) for digital channels, prompting an increase in investment in these areas. Additionally, it highlighted regions with suboptimal performance, allowing for strategic reallocation of financial resources.
Impact
- $14M increase in profits across five regions and two major brands.
- MMx models scaled globally across 30 markets in 18 geographies, driving consistent insights and execution.
Business Impact
-
$14M
in Profits
-
30
Markets Scaled with MMX Modeling
Let’s move from data to decisions together. Talk to us.

The firm's name is derived from the statistical terms "Mu" and "Sigma," which symbolize a
probability distribution's mean and standard deviation, respectively.