Catalyst Project
Machine Learning Engine for Product and Sales Management through Digital Channel
Prepaid customers belong to and switch among three main propensity profiles: potential churn, potential growth and stable. According to this, actions* and offers** can be recommended directly and automatically to customers to generate value or avoid negative impact. Machine learning techniques can be applied to understand and predict three main things:
- What profile group a customer belongs to;
- Which product can be offered to maximize customer value or what action the customer can “ask the system” to do in order to avoid negative impact;
- Apply reinforced learning techniques that will help select the best set of actions or offers a customer’s segment should get as recommendations.
The customer’s interaction goes through the CSP’s app using a cognitive agent that interacts with the MLe (Machine Learning Engine) to orchestrate the customer’s complex interaction with: ML results, a marketing management service, product and customer catalogs and a RPA service that assists in the execution of actions.
* Examples of actions: activate CSP’s music app or an unused benefit, change current product mix…
** Examples of products: up-sell to get more data, migrate to a specific postpaid plan, cross-sell wireline Internet, buy a TV or gaming console…
Articles & blogs
Champions
Participants
Video
Machine Learning Engine for Product and Sales Management through Digital Channel
Catalyst: Machine Learning Engine for Product and Sales Management through Digital Channel