Predictive Analytics

What is Predictive Analytics?

Predictive analytics in short is a practice that “encompasses a variety of statistical techniques from modeling, machine learning, and data mining that analyze current and historical facts to make predictions about future, or otherwise unknown, events.”

It works simple on the surface, you input the data and then one of those models outputs predictions. However, if you have been working in analytics for years and have to deal with vast amount of data that gets generated on a high velocity and volume each and every day, you know it’s not easy to drive insights from these sprawling and disparate data sets based on manual wrangling or human intuition - there is a lot beyond that.

Predictive Analytics for Customer Success

Customer Success, as a new emerging profession in the SaaS environment which bears the mission for helping reduce SaaS churn rate and increase customers retention, appears to be the new fortress for predictive analytics in most recent years.

The primary goal of a Customer Success solution is to improve the productivity and effectiveness of the CSM team by alerting them to customers who need attention (as well as informing why and suggesting what to do about it). Most solutions generate alerts for accounts at risk of churning, but some can also predict which customers are likely to upgrade (up/cross-sell) or likely to convert from trial or freemium to a paid account.

One key differentiator of Natero versus other CSM solutions that rely solely on rule based system is that we have built a predictive analytics element into the solution. In addition to providing the basic analytics and alerting capabilities others provide, we are also providing “data science as a service” -- building advanced self-evolving machine learning models which can be applied to each customer's unique business scenarios.

How it Works?

At Natero, we use state-of-the-art machine learning technology to build predictive models and algorithms that apply to SaaS vendor’s unique scenarios and types of data to proactively alert them to customers at-risk, as well as those that represent sales opportunities.

How we do this, in general, is to look at vast amount of historical data of your customers, and build models around these data to identify common behaviors, patterns, or attributes of those customers (e.g. size of customer, activity level, feature usage, invoice history) who have churned, converted, or expanded; and then use these models to predict if any of the above event is likely to happen to a current customer and how reliable that prediction is.

In other words, these advanced machine learning models analyze hundreds or thousands of factors to determine which are the most relevant indicators of churn, expansion, conversion, etc. Normally these factors are not apparent or easily captured by a simple business rule.  

This process involves concepts such as “data preprocessing” -- the way of converting raw data to meaningful data sets; “feature selection” -- the way of finding the most useful feature in module building; “data training” -- a set of data used to fit a model that can be used to predict future business outcomes; “data labeling” - which makes data more readable and thus allows careful and meaningful use of data; as well as “feature ranking” -- the method of deciding which features are most useful to get better prediction in a model, etc..

When to Use it?

Predictive analytics offers greater accuracy than those that rely on intuition or guesswork, when it comes to predicting future customer actions and business outcomes. However, it is useful only when enough historic customer data has been gathered and consumed; also these data need to be properly labeled for e.g. "churn", "upgrade" and "trial conversion" etc..

As more data comes in, Natero’s machine learning models automatically adjust for changes in customer behavior and become more accurate in their ability to predict customer actions - with absolutely no effort required from you.

 

Learn how Natero builds predictive models.

Have more questions? Submit a request

Comments