Abstract  

We Sell Everything in Software (WSES) Inc. dealt with innovative off-the-shelf products and had a high growth trajectory. They had a healthy pipeline and their annual marketing spend of $500 million was allocated on the basis of conversion probabilities at flat 6% of the expected sales value. WSES did not have a structured approach for calculating sales conversion probabilities – it followed the gut feelings of its marketing and sales team. Therefore, it incurred high marketing cost including travel costs, client visits, time spent by the sales team/technical experts/support staff, and logistics costs.

 Jack, the CEO of WSES, was worried that despite having such a huge expenditure list, the sales conversion possibilities based on the pipeline was at best an ancillary information, as there was no substance in justifying the ‘‘gut feeling’’. Thus, Jack was not too convinced of this method and was exploring multiple options with Ben, Vice President, Marketing. Considering that WSES pursued around 1,000 opportunities every year, which they either won or lost, Jack felt the need to have these data that they had collated over the years validated and thereby interpreted. Thus, WSES engaged Mark with a Ph.D. in statistics to understand if they could determine a structured approach to compute conversion probabilities and devise an optimal way to allocate marketing spend such that even with significant reduction in marketing spend, the objectives on expected values on sales and profits were met.

Learning Objective  

The case focuses on determining a structured approach to compute conversion probabilities of sales leads and seeks to devise an optimal way to allocate marketing budgets. The case discusses the use of discriminant analysis to segregate opportunities into win and loss deals. It may be used for understanding integration of predictive analytics with prescriptive analytics: probabilities that are computed using discriminant analysis are used as an input to compute the expected value of sales and profit. This is an important factor to determine optimality.