To enable decision makers take actions proactively and hence improve customer satisfaction.

In one the largest technology company, the Projects were reviewed on a periodic basis and flagged as Red or green depending on the status of the project. However, this helped take corrective measures but could not prevent issues from happening. Hence they wanted to build a framework to predict which projects will be flagged as RED over the next quarter and the causes of the same so that an action can be taken proactively.

We collected the demographics of the projects  and the historical data on each of the parameters on which the projects were rated as the data was at disparate sources. Supervised Machine learning technique was used to create an algorithm which predicted the probability of the project becoming RED and the rules which caused this.

The model helped reduce the number of red flagged projects and as a result of this, the Customer Satisfaction ratio increased significantly. The proactive actions resulted in saving close to $250,000 .