Have you ever worked in the eCommerce industry and got a chance to work on raw data like orders received, orders returned by the customers or details of the customers? If yes and you always thought of how this data affected marketing strategies, I have the perfect answer for you. Customer Analytics – a branch of analytics focusing on factors like Customer Lifetime Value and Centricity.
To be honest, it has a lot to do with marketing jargon and statistical models. However, the intention to write this blog is more to do with which topics to focus. I will be writing about Predictive Analysis and focusing on how customer data is analyzed to derive insights.
There are roughly two background settings to pay attention to – Contractual and Non- Contractual setting.
A contractual setting is a scenario where the attrition of the customers is known to the firm most precisely the time when they stopped participating in doing the transaction with the firm while in the non-contractual setting, no data is available on customer attrition which means distinguishing between the customers who have left and who are new is not possible in this case.
What can we determine using customer data?
- Which customers are more likely stay active in future?
- The transaction value expected from above customers if they stay
- Customer Lifetime Value
To evaluate point 1 stated above, we shall focus first on the analysis technique called RFM (Recency, Frequency and Monitory Gain). This technique has been widely used by the marketing teams in organizations to identify the best customers on the basis of the three factors mentioned below:
- Recency – The duration of time when they were last active
- Frequency – The number of times they have made the transaction with the firm.
- Monitory Gain – The amount of money spent by the customer
The ongoing debate is why the model is named RFM than anything else. Is there any particular order or rank being followed here. The answer is Yes, they follow the order due to dependence. If the Recency is in place, we can be sure that the customer is active and chances are high of him becoming the best customer as compared to someone who has been absent in previous periods. Once, the recency factor is met, the other factors can be considered.
RFM model consists of parameter-based evaluation where the customer is given value or parameter ranging from 1-5, 5 being the highest. The database is then searched for the customers with RFM value as (5,5,5) and marketing is designed targeting these customers for high returning value.
Classification of customers on the basis of relationship
The customer data can be used to gauge customer activity in terms of the transaction. The activity could be continuous or discrete in nature. Continuous activities can be understood with the example of grocery shopping, doctor visits and all these activities come under non- contractual segment. Non-contractual Discrete activities can vary from charity events and conferences.
The Continuous and contractual set of activities vary from Credit cards to continuity programs. The Discrete and contractual set of activities can be understood with subscription programs.
Customer Lifetime value – Why is this useful?
It helps in calculating the company net profit from a customer. This strategy helps the organization in calculating the amount of money they want to spend on attracting new customers by aligning marketing campaigns and the amount of transaction expected from repeat business from existing customers.
The companies here calculate the cost of acquiring the customer, cost of marketing efforts taken for that customer and the net profit after the customer has made a transaction. Customer lifetime value is one topic which needs more detailed approach. I shall cover a detailed piece soon targeting the most searched questions.
