Reducing Customer Churn in Data Science Fashion

M. Ahmed Tayib
11 December 2020
5 Minutes Read

Solving the decade-old question with data-science

Old Problem in a New Time

Customer churn for any business is a nightmare, no matter the times we live in. However, this has only been intensified by today’s online world. We live in a time when everything is online, and everything is subscription-based. You wanna listen to music, well subscribe to Spotify or Itunes. You wanna watch TV shows; better subscribe to Netflix or Hulu, you wanna watch sports; subscribe to another sports service like Espn or SkySports. My point is we live in a fast world where everything we do is online; it’s the new norm.

Once losing a customer meant losing some potential future sales and revenue from that customer. For subscription-based companies, the scenario has changed. Losing customers means losing a steady stream of guaranteed monthly revenue from those customers. The same case can also be said about the e-commerce industry as well. Since we see a surge in online shopping, the situation is more pressing for e-commerce. If you are losing your customers, you would wanna know the reasons behind it so that you can address them and be more competitive. Also, you would wanna do anything to reduce future churn cases as much as possible.

New Era New Ways

Everyone has a different opinion on how to reduce churn. But opinion can take you only so far. The most effective way is to let your data speak and spill out the ways to reduce churn cases. We have all gotten accustomed to the idea of decision making using data in today’s world. So let’s apply the same thing here as well.

The most effective way to reduce churn is to prevent it before it happens. Suppose you know a customer/user will churn next month and know the possible reasons why that customer is about to churn. In that case, you can reach out to that customer and offer some coupons or engage him/her to retain a customer and prevent a possible churn. In other words, Churn Prediction.

Don’t let Churn Prediction intimidate you. I will show you one of the easiest ways to do that, even without writing a single line of code. To do that, we need 3 data; Sales Data, Customer Data, and Product Data. We will do that using a platform called Enhencer. Just create a free account here, and we are ready to do it.

The Data Science Fashion

Enhencer is a self-service machine learning platform that lets users create models like churn predictions without any prior knowledge of programming. First, we need to upload the data to the platform using the connector that suits us the most.

propensity-connect

Let’s upload these three data using CSV, and when uploading is finished, let’s click the Explore button.

Remember I told you now to get intimidated by Churn Prediction; here is why: once you upload the data Enhencer does the rest on its own. Enhencer:

  1. Does Feature Engineering automatically on the data
  2. Trains multiple Machine Learning Models on the processed data and chooses the model with the best performances.
  3. Provide you with the churn predictions for all your customers in the customer data.

Here is how the predictions are presented to you:

propensity-connect

Let me break down the most important elements from the dashboard for you. The data we uploaded had a total of 2474 customers; among them, 32% churned in the past.

Enhencer also segments the customers based on their past behavior and their likelihood to churn in the future. Segment 1 contains the customers with the highest probability to churn, meaning these 19 are the customer most risky. There is a 77.8% chance that these customers will churn in the next three months.

propensity-connect

Enhencer provides the behavior of these customers as well. These 19 customers who are very, very likely to churn have a few common behaviors. In the past, these 19 customers bought products from only one product category. They have been inactive for at least 55 days. They gave a product rating of below 1 and are more than 27 years old.

This is a very powerful insight. This means you can reach out to these 19 customers with some marketing campaigns that are very much personalized based on their behavior. In other words, offer them something they can’t refuse, and you will have a very high chance to retain them and prevent them from becoming churn.

Similarly, if you reach out to customers from, let’s say, top 5 segments and organize marketing campaigns personalized based on their behavior, this will bring down churn significantly in the long run.

Not Intimidating now, Is It?

See, I told you you won’t have to write a single line of code. All you need to do is upload the three data to the platform. Another thing I want to emphasize is the customer segment behaviors. You might be saying, how can I create these behaviors in the data. Well, the thing is you don’t have to. Those behaviors are extracted by Enhencer automatically from the data that you upload. It uses Machine Learning Algorithm to do so.

Once the predictions are provided, just download the customer lists from any number of segments. You know which customers are going to churn and know the reasons and behaviors behind the churns. Simply reach out to them using an offer, personalized marketing campaign, or whatever is the policy in your company. In this way, you are sure to reduce the churn rate significantly with time.

Author photo
Written By M. Ahmed Tayib
Discover More Articles Like this