Benefit Prediction of Buying Customer Relationship Management Features of Pipedrive
Name
Alo Aasmäe
Abstract
People that use software such as CRM applications, which enables to track relationships with their clients, have always been interested in knowing the future prospects of their clients, especially when additional value could be offered to the client. Applying process mining techniques in combination with adversarial model training to make these predictions have been scarcely done in the context of CRM applications to predict user behavior. By defining an event that is assumed to have an effect on user actions, it is possible to split a user’s event logs into two parts based on the timestamp of this event, enabling to create a mapping between the user’s actions before and after. This thesis applies both process mining and machine learning on event logs to predict future user
actions based on previous user behavior. Here we show that this approach is viable in a business context by using event logs extracted from Pipedrive users and that the solution could provide value in marketing scenarios. The result is an automated way of predicting user metrics for custom "what-if" scenarios, provided that sufficient event logs are present. The effect of this is potentially beneficial for both the end-user and employee of Pipedrive, by enabling a way for a user to see the effects of further investment and providing a better direction for allocating resources in marketing.
actions based on previous user behavior. Here we show that this approach is viable in a business context by using event logs extracted from Pipedrive users and that the solution could provide value in marketing scenarios. The result is an automated way of predicting user metrics for custom "what-if" scenarios, provided that sufficient event logs are present. The effect of this is potentially beneficial for both the end-user and employee of Pipedrive, by enabling a way for a user to see the effects of further investment and providing a better direction for allocating resources in marketing.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Gamal Elkoumy, Roland Kriibi
Defence year
2022