Churn prediction made easy

The ability to predict that a particular customer is at a high risk of churning, while there is still time to do something about it, represents a huge additional potential revenue for every business.

Now, thanks to the Predictive Analytics platform we offer, predicting churn is no longer exclusive to big players that can afford to hire teams of data scientists. Everyone can use it, it is easy!


Predictive analytics is the use of statistics, machine learning and data mining to analyze current and historical facts to make predictions about future events. 

Big data and analytics have emerged as key assets in enabling companies to differentiate themselves from their competitors and to drive business success. However, whilst many organizations have mastered data collection, they fail to operationalize the value of data - and opportunities for business growth are being missed.

There are three possible strategies businesses can use to generate more revenue: acquire more customers, upsell existing customers, or increase customer retention. All the efforts made as part of one of the strategies have a cost, and what we’re ultimately interested in is the return on investment: the ratio between the extra revenue that results from these efforts and their cost.

If you could predict in advance which customers are at risk of leaving, you could reduce customer retention efforts.

The software service we offer, delivers:

  • Easy but accurate way to predict who is going to churn
    Using our Lifted Churn method. We just add a column to your data saying: Churn Yes/No
  • Business solution
    So no IT- or technical project, no analytical skills required, just use it
  • Short time to market
    No long projects, no up-front investment, just use it!
  • Highly automated data preparation
    Based on experience in the past
  • Strong security
    Your data and network activity are encrypted. Nobody but you has access.
  • High performance
    Carry out complex predictive analyses on large data.


How does predicting churn work?
By providing historical data to the service, you enable it to learn and train a model that fits your business. After the model has learned this patterns, you will be able to provide recent history for a set of customers and will receive a prediction identifying the subset of customers that have high risk of leaving your business.

Using this service
Get started predicting churn for your business by following simple steps:

Provide historical data to the service. This data needs to cover the interaction between customers and your business. (e.g. details of purchases made by customers) for a consecutive period of time based on your churn problem description. Using this data, the service will train various models to predict future churn risk for your customers.
Try your new models! You will need to provide similar history data to the one you used for learning. However, this data should cover a different set of customers. In return, our service will add two new columns to the dataset: one saying "Churn: Yes/No" and a column containing the probability score for each customers - the higher the score, the higher the probability is that the customer will leave.


We love to make complex things as simple as possible.

Predictive Analytics is no longer exclusive to big players that can afford to hire teams of data scientists anymore! We put a lot of predictive analytics technology under the hood, all with the goal to make the software usability ultimately easy. 

The need of the hour for the organizations today is not only to know what is happening to their business, but also be able to predict what is about to happen.
Organizations need to make big data predictions an integral part of their everyday workflows, which means running analytics processes continuously and easily. Our prediction software is exactly enabling that.