Conference proceeding
Building a Better Benchmark: Predicting Effects of Shopper Marketing on Sales
2022 Systems and Information Engineering Design Symposium (SIEDS), pp 363-368
28 Apr 2022
Abstract
Companies with wide product portfolios and multiple retail channels often have difficulty quantifying the sales impact of marketing programs due to the large number of factors that potentially influence sales. The amount of data and complex modeling necessary to get such a promotional model off the ground creates a challenge, especially for firms executing a wide variety of promotions across many retailers. This challenge can stunt any efforts to make data-driven decisions regarding marketing spending. Our work explores marketing program data from a national consumer packaged goods (CPG) manufacturer and related product sales data from one of its retail partners. We build two separate models that provide a measure of incremental sales attributable to marketing programs at the brand level. The findings show that under certain conditions, organizations can achieve a useful promotional sales model with modest data inputs. Applying this approach, organizations can gain insights into the sales impact of their marketing spending, especially if they incorporate partner data, limit data streams and features, and incorporate program tactics. Our models can be used for descriptive as well as predictive analysis, thus allowing a CPG company to improve decision making that relies on forecasts of future sales.
Metrics
1 Record Views
Details
- Title
- Building a Better Benchmark: Predicting Effects of Shopper Marketing on Sales
- Creators
- Patrick Corbett - University of VirginiaNicholas Keeley - University of VirginiaGabriella Belmarez - University of VirginiaF. Will Blickle - University of VirginiaOliver Schaer - University of Virginia
- Publication Details
- 2022 Systems and Information Engineering Design Symposium (SIEDS), pp 363-368
- Publisher
- IEEE
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Decision Sciences (and Management Information Systems)
- Scopus ID
- 2-s2.0-85134298697
- Other Identifier
- 991021862273904721