fbpx

Your support helps keep our site running! We might earn a small referral fee when you purchase from links in this post, at no extra cost to you, which we REALLY appreciate. All opinions remain our own as always.

Detecting Customer Un-Enrollment Before It Happens!

Mark O. | At-Home Healthcare Service

Detecting Customer Un-Enrollment Before It Happens!

Mark O. | At-Home Healthcare Service

about this project:

A budding At-Home Healthcare Service had a great vision to help bring healthcare to the homes of the elderly, however they were struggling to keep customers enrolled in their services. Discovering why customers were leaving would allow them to provide better care to their patients and offer incentives to prevent customer churn.

Mark O Deck Slide Deck Churn | OWLLytics.com

Results

By utilizing a machine-learning approach to analyze their patients’ demographics, the client was able to launch targeted incentive campaigns to keep their customers satisfied and enrolled in their service.

%

Customer Churn Rate (Before)

%

Customer Churn Rate (After)

Monthly Revenue Recovered (USD$)

CHALLENGES

THE CLIENT NEEDED A SYSTEM IN PLACE TO IDENTY CUSTOMERS BEFORE UNENROLLMENT, WITH HOPES TO PROVIDE BETTER SERVICE & GROW AS AN ORGANIZATION.

With over 25,000 customers, the client had a large amount of data, but little understanding of why a their customers were cancelling their service within the first three months. 

Although they had a general understanding of the problems, they were struggling to identify these factors in customers BEFORE they actually cancelled their enrollment.

As a healthcare provider, the company needed to abide by strict legal regulations regarding the management of their data, which they were currently managing in Microsoft Excel spreadsheets.

This added an extra layer of importance to their challenge: Creating a way for the company to detect unenrollment beforehand, while maintaining data privacy.

Mark S Happy Frown | OWLLytics.com

CORRELATION EXAMINATION 

SOLUTION

OWLLYTICS BUILT A MACHINE-LEARNING MODEL THAT COULD ACCURATELY IDENTIFY CLIENTS AT RISK FOR UN-ENROLLING, GREATLY REDUCING CUSTOMER CHURN.

OWLLytics was able to build a specific machine-learning model for the company to see which customers had similar ‘unenrollment’ risk factors, classifying them as customers who were ‘at risk to unenroll,’ with a 99% certainty.

The company had a comprehensive data set of over 25,000 clients, with 28 different variables (demographics) which could be used to look for patterns in customers who dis-enrolled. 

➡ Once we identified at-risk customers, the company reached out to better understand their needs and see what services could be improved. 

➡ Changes were implemented with a goal of customer retention, including offering additional benefits specifically to ‘churn-risk’ customers.

➡ Going one step further, we created the model to be easily updated and re-usable, to be a consistent benefit to the company without the need for data-trained staff.

RESULTS

USING THE PREDICTION MODEL HELPED THE COMPANY TO INCREASE THEIR ANNUAL REVENUE BY OVER $4.0 Million.

Mark S Slide Deck In Computer | OWLLytics.com

➡ The company launched a customer satisfaction campaign specifically geared towards their churn-risk customers.

➡ They increased the satisfaction rate among ALL customers by addressing issues discovered during this campaign.

In addition to upgrading their currently services to better match customer needs and expectations, they also offered targeted incentives to at-risk customers in order to prevent unenrollment. 

➡ The company was able to reduce their customer churn rate from 6.99% down to 3.57% .

➡ The revenue gained by retaining these additional customers was $4,000,000.00 per year.

➡ The additional revenue provided powerful opportunities for the company to grow and maintain long-term customer satisfaction.

Let’s Start a Conversation

8 + 7 =

Let’s Start a Conversation

8 + 3 =