What are Predictive Analytics?

Predictive analytics is a form of advanced analytics that determines what is likely to happen based on historical data using machine learning. Historical data that comprises the bulk of descriptive and diagnostic analytics is used as the basis of building predictive analytics models. Predictive analytics helps companies address use cases such as:

• Predicting maintenance issues and part breakdown in machines
• Determining credit risk and identifying potential fraud
• Predicting and avoiding customer churn by identifying signs of customer dissatisfaction

How Do You Get Started with Predictive Analytics?

At the outset of any predictive analytics build, three core elements need to be established:

• Identify a problem to solve
• Define what is you want to predict
• State what you will achieve by doing so

To start, you should collect existing data, organize it in a useful way to allow for data modeling, cleanse your data and review overall quality, and finally determine your modeling objective.

While modeling takes up the spotlight in predictive analytics, data prep is a crucial step that needs to happen first. Therefore, organizations with a rock-solid foundation in descriptive and diagnostic analytics are better equipped to handle predictive analytics.

Simply put, the time and effort to prepare, transform, and ensure data quality for retrospective reporting has already taken place. The groundwork should be relatively well laid to quickly identify and leverage data for the modeling phase.

We always encourage customers with well-defined KPIs and business logic in a specific business reporting area (think sales reporting for example) to use that as the first predictive analytics use case. The goal is to derive value quickly, and there is no better place to start than an area where you know data is well defined and of high quality.

Continue with Ember as we take a look at the fourth type of Business Analytics: Prescriptive Analytics.