kaos Discover kaos, our new opensource platform for simplifying machine learning deployment! Learn more

Customer Segmentation

Data-driven decisions drive marketing efforts toward more customer-centric marketing approaches. Enabling our clients to know their customers, we help them to back their marketing strategies.

By profiling the customers and calculating the Customer Lifetime Value, the marketing team can evaluate its effort and focus on acquiring the right customers, which is essential for the success of the business in the long run. Customer segmentation is a common activity that helps companies to understand and profile their customers. Marketing teams use customer segmentation to target customers effectively and appropriately.

With carefully analyzed data, our clients can:

  • personalize the marketing campaigns
  • enhance customer experience and engagement
  • optimize marketing channels
  • optimize the quality of the online content
  • focus on acquiring the right customers who are the primary factor in the success of our clients' business.

How does it work?

We used the Google Analytics API to extract various tables of data based on the selected features and metrics. The tables were extracted in a way that one can join them later based on given primary keys. The data has been cleaned and prepared before applying unsupervised learning. After grouping the customers using more than 35 features, the marketing team can create different groups of audiences in Google Analytics and personalize marketing campaigns. The principal component analysis (PCA) was applied to reduce the dimensions of the data, and then the K-means method was used to cluster the customers.

In marketing, Customer Lifetime Value (CLV) is used as one of the most important metrics to measure the success of the business. CLV shows how much money the business owner should be spending on acquiring customers. In other words, CLV is showing how much value customers will bring to the business in the long run. There is no straightforward method to calculate CLV for all cases. However, researchers developed a simple technique to estimate CLV using a simple set of available data. The simplified technique uses “Recency, Frequency and Monetary Value” RFM-framework to measure the value of each customer. The RFM data can be calculated from the database easily.

age T represents the age of the customer in whatever time units chosen (weekly here). This is equal to the duration between a customer's first purchase and the end of the period under study.

recency: represents the age of the customer when they made their most recent purchases. This is equal to the duration between a customer’s first purchase and their latest purchase. (Thus if they have made only 1 purchase, the recency is 0.) Here, the recency unit is weeks

frequency represents the number of repeat purchases the customer has made. This means that it’s one less than the total number of purchases. It’s the count of time periods the customer had a purchase in. So if using days as units, then it’s the count of days the customer had a purchase on. Here, we calculated the frequency as the count of the days -1.

monetary value represents the average value of a given customer’s purchases. This is equal to the sum of all a customer’s purchases divided by the total number of purchases. Note that the denominator here is different than the frequency described above.

To analyze data, we used the “lifetime” framework, which is implemented in python. The “lifetime” package provides flexibility and productivity by implementing several models to fit the data and visualize the results quickly.

Future steps

In future work, many successful recommendation systems and advanced customer relationship management tasks such as creating leads and converting them to opportunities can be built on top customer segmentation.


Meet the team

Find out more about our unique culture and diverse team.

Read more   ➞