HomeNewsGuidesReadsPodcastsTRANSFIN. EOD
  1. Reads
  2. Lite

Artificial Intelligence for Better Customer Service: Reward Loyalty and Manage Churn

Data Scientist, DirecTech Labs
Jun 2, 2018 4:57 AM 3 min read

It is an accepted reality of doing business that customers come and go. Loyal customers might stay with you longer than new customers, but sooner or later they will churn. The idea then is to keep on increasing the tenure of each customer with the right customer engagement strategies. It is key to use Machine Learning for understanding your customers and reducing attrition.


Let’s look at:

  • What is churn?
  • How does it impact the top-line of a company?
  • How can you predict who is going to churn?
  • What should you do to stop the customers from churning?


What is Churn?


Churn or attrition rate is the number of customers who have used your products or services in the past but are not going to continue using them in the immediate future.


For different businesses, it can be defined in various ways:


Retail – Customers who aren’t going to purchase anything in next 2 months could be considered as churn (or inactive, dormant etc.)

Telecom – Customers who have raised the request to disconnect services or those who have already disconnected their connections 

Retail Banking – Customers who have not used your services over a certain time-period and / or are not keeping the threshold balance


How you calculate churn can vary from industry to industry. It also varies for different segments in the same company. Make sure you understand your customers at both the micro and macro segments.


How Does Churn Impact the Top-Line of A Company?


Consider this,


A company ABC has more than $100 million in annual revenues, with 100,000 active customers. Now this broadly translates into each customer providing $1000 in revenue.


Let’s assume the acquisition cost for each customer to be $100.


Assuming 1,000 customers leave at the start of the year means you are looking at losses of 1000 * 1000, which is $1,000,000.


In order to acquire the same customers, you will have to once again spend 100 * 1000 = $100,000


Hence, the Net loss with 1% churn rate is $1,100,000


I hope, you have my attention now!


How Can You Predict Who Is Going to Churn?


If you have only 10 customers – you might already know who is going to churn because you have the most advanced algorithm in your body i.e. your brain.


If your company interacts with 1000s of customers, you need technologically advanced processes which can identify ‘at-risk’ customers at the right time. That is where Artificial Intelligence becomes handy.

Artificial Intelligence for Better Customer Service: Reward Loyalty and Manage Churn

Based on the trends and patterns in historical data, one can gauge with certain accuracy who is going to churn and more importantly, what are the main factors driving that churn?


If Customer is King, then Data is Queen


How better your company can predict the churn depends on:


  • Number of touch points (transaction data, demographics, call center, service requests, complaints) your customers have with you
  • Duration for which these touch points have been stored
  • Quality of all the data points. Ensuring that no incorrect data is being stored


In my previous experience, we can predict with 80-90% accuracy who is going to churn in the near future.


What Should You Do to Stop Customers From Churning?


You have to understand your customers both at a micro and macro level.


You should have answers to the below questions. And based on your answers, you can devise strategies to increase engagement for each customer.


Personalization is the key word in today’s business.


  1. Which customer segment is generating the most revenue for you – Loyal Customers?
  2. Which customer segment is very new to your company – New Joiners?
  3. Which customer segment is composed of infrequent buyers – Dormant Customers?


Answers to the above questions will help you devise the right strategies for each customer.


For instance:


Loyal Customers – These customers are very critical for your business and they deserve your urgent attention. Target them with most lucrative offers and make sure they have a wonderful product experience. Remember, these are loyal customers, who often bring newer customers.


New Joiners – New customers should be treated differently. Help them enhance their product experience. Distributing user manuals/videos could help them understand your products.


Dormant Customers – Identify if customers using a specific product are experiencing more churn. Introduce them to your best products, instead.


Originally Published on Medium


(We are now on your favorite messaging app – WhatsApp. We strongly recommend you SUBSCRIBE to start receiving your Fresh, Homegrown and Handpicked News Feed.)