Business Objective
A Manufacturing company would like to improve the conversion rate of sales leads by identifying promising ones using machine learning and increasing sales revenue. Businesses would like to implement a lead scoring system in CRM to identify leads with a high probability of conversion.
Solution
Lead Information from the internal CRM system and 3rd party market information is used. Customer-specific information such as revenue, consumer base, and geographical location were added. All categorical variables were encoded, and geographical variables like zip code were substituted with relevant variables such as population and area.
Appropriate features were then selected to be used in the model. A supervised learning model (XG Boost) was used to predict the probability of a lead being converted.
The probabilities were presented to the sales team through a web app that could take feedback from the sales team to retrain the model.
Impact
The model had an accuracy of 74% and significantly improved lead conversion rates for the sales team.
** The client has chosen AIDAS Analytics Office (subscription-based pay-as-you-go) services to run business analytics projects.
