BUSINESS INSIGHTS

Oct 17, 2017

Maximize Your Customer Retention by Predicting Customer Churn

Colby Ford Posted by Colby Ford

"Increasing customer retention rates by 5% increases profits by 25% to 95%." - Harvard Business Review[1]

Reading this statistic in Frederick F. Reichheld and Phil Schefter's "The Economics of E-Loyalty" article from the Harvard Business Review, I was surprised that such a minor tweak in retention could have such a major impact on profits. The economic experts went on to explain that loyal customers buy more and refer new customers, solidifying the importance of long-term customer relationships.

customer churn.png

While we know winning a new customer's business is difficult and costly, keeping long-established clients has its own set of challenges. Predicting and understanding customer churn take some of the guesswork out of customer retention. By studying behaviors of clients who have churned in the past, and looking at current customers that are behaving similarly, we can take action to positively affect their retention. This can be achieved by establishing a targeted plan that nurtures loyalty in every one of your customers.

Key Benefits, Facts, & Insights

  • Boost Profits: Selling to existing customers is easier and more cost effective rather than selling to new ones.
  • Retain More Customers: Proactively launch campaigns and strategies to abate customer attrition.
  • Win Back Business: Identify the root causes as to why customers leave and establish re-acquisition strategies.
  • Avert Loss: Customer loss is substantial, long-reaching, and can impact everything from revenue to opportunity for competition.

Where to Begin...

Whether your customers are people or businesses, and no matter if you're selling products or services, the problem is still the same: Keeping current customers while continuing to grow with new ones. This means understanding who is no longer a buyer and why they've chosen to leave. Once you have the answers, you can work on your retention practices to prevent similar losses in the future.

Knowing who is no longer purchasing from you is easy. The mystery lies in knowing which active clients you may lose. Predicting which of current customers are exhibiting behavior like your "previously-churned" customers is the best way to tackle the problem. However, doing this by hand would prove to be highly complex. Using the power of Machine Learning and Microsoft Azure, your historical customer data will be put to work to accurately predict future churn.

We're using our Customer Churn Solution to help organizations utilize their historical customer data to target at-risk customers and find opportunities to bring back those who've left.

  • The first step is to define what "churn" actually means for your organization. Often, "churn" is defined by a period of inactivity. Additionally, defining your success criteria is very important to this step.
  • Once we help you gather sufficient historical data to make a satisfactory prediction, we can then begin creating the predictive model. This also includes model validation and tuning to get the most accurate churn prediction from your data.
  • As a last step, we will consult with you around implementing this churn prediction into your organization, both from a data architecture front and reporting/training standpoints, too.

Putting the Predictions to Use

Our goal is to give your company a targeted set of actions to improve customer retention. Once an accurate model has been trained, all of your past and present customers will be run through the predictive model. This could give one of three results. See the table below:

Churn_Table.pngIf the prediction matches the actual status of the customer, then there is no action necessary. However, if an active customer is predicted to be a churned customer, this will imply that this customer is at-risk of churning. Alternatively, if a customer is currently inactive (meaning that they previously churned), but the prediction is that they are an active customer, this may indicate that they are a good target with which to attempt to reinstate business.

churn.png

BlueGranite's Customer Churn prediction also enhances reporting by providing account managers with a list of clients to target with retention efforts, and identifying the best former prospects to try and win back. These crucial insights maximize customer preservation and accurately pinpoint marketing efforts.

Want to Learn More?

Interested in learning more about customer churn, but not sure where to go? Here are a few more resources to point you in the right direction:

References:
[1] Harvard Business Review - https://hbr.org/2014/10/the-value-of-keeping-the-right-customers
Predicting Customer Churn
Colby Ford

About The Author

Colby Ford

Dr. Colby Ford is a Data Scientist at BlueGranite. Coming from a background in mathematics, statistics, and computational biology, he combines this expertise to bring AI to everyone. Using R and Python, he puts Machine Learning to work to gain insight from data. Outside of BlueGranite, Colby is an avid genomics researcher. Check out Colby’s website at www.colbyford.com.

Latest Posts

Predicting Customer Churn