What is customer lifetime value modeling?
Posted: Sat Dec 21, 2024 5:24 am
I'll take an energy company as an example:
Renewals: These people remain customers and consciously renew their contract or purchase a product from the company again.
Sleepers: These customers are not renewing or terminating their contract. They are also called sleeping dogs. They let the contract continue and remain customers, sometimes they do this unconsciously. This group usually brings in quite a bit of money, sometimes it is smarter not to run campaigns with these customers (don't wake sleeping dogs).
Churners: These customers are most likely to cancel their contract with the company or switch to another supplier.
With data from the past of renewals, sleepers and churners we train an algorithm. Then it can learn from the patterns from the past to predict what will happen in the future, on an individual level. In this way, a certain chance of churn, a chance of retention (the permanent customers) and a chance of sleepers are predicted for all customers. With these insights we now know for which customers we should do our best to retain them, which customers are most likely to continue their product/service themselves and which customers will go 'asleep'.
2. Customer lifetime value modelling
Now we probably know what a customer is planning. What now? The next step is to calculate the value of the customer. You can achieve this by using customer lifetime value modeling. Here you map out what the value of the customer is.
Customer lifetime value modelling (also called CLV modelling) is a technique that calculates what the customer will yield and cost. This econometric approach provides insights to determine whether a customer is worth investing a certain marketing budget. In some cases, it may even be better to let a customer go, because a marketing campaign with those customers costs more money than it yields. CLV modelling helps to achieve the ultimate goal. Using budget and laos telegram data resources with maximum impact to create more value.
For example, I sometimes see that customers have a telephone subscription or energy contract with minimal consumption and minimal profit margins. Expensive marketing campaigns are then conducted on this. Smart if you have a well-founded expectation that these customers will bring in more turnover or margin in the coming months or years. Not smart if the expectation is that this turnover will remain minimal, and the margin remains marginal or even becomes negative. This is exactly what you are trying to achieve with CLV modelling. Mapping the current and future value of customers.
Stay critical and work with segments
When using CLV modelling in the B2B sector, be careful with the Pareto distribution (20% of your customers represent 80% of your turnover). For example, in the telephony sector. A company with 20 telephone numbers will generally have more value than a self-employed person with 1 telephone number. However, this does not mean that all self-employed people together have no value. There are also differences between customers within this group. You can solve this by working with segments and calculating the value in relation to each other within them. You can then better manage (the potential) value within the segments. This prevents you from no longer approaching a large part of your customers.
Renewals: These people remain customers and consciously renew their contract or purchase a product from the company again.
Sleepers: These customers are not renewing or terminating their contract. They are also called sleeping dogs. They let the contract continue and remain customers, sometimes they do this unconsciously. This group usually brings in quite a bit of money, sometimes it is smarter not to run campaigns with these customers (don't wake sleeping dogs).
Churners: These customers are most likely to cancel their contract with the company or switch to another supplier.
With data from the past of renewals, sleepers and churners we train an algorithm. Then it can learn from the patterns from the past to predict what will happen in the future, on an individual level. In this way, a certain chance of churn, a chance of retention (the permanent customers) and a chance of sleepers are predicted for all customers. With these insights we now know for which customers we should do our best to retain them, which customers are most likely to continue their product/service themselves and which customers will go 'asleep'.
2. Customer lifetime value modelling
Now we probably know what a customer is planning. What now? The next step is to calculate the value of the customer. You can achieve this by using customer lifetime value modeling. Here you map out what the value of the customer is.
Customer lifetime value modelling (also called CLV modelling) is a technique that calculates what the customer will yield and cost. This econometric approach provides insights to determine whether a customer is worth investing a certain marketing budget. In some cases, it may even be better to let a customer go, because a marketing campaign with those customers costs more money than it yields. CLV modelling helps to achieve the ultimate goal. Using budget and laos telegram data resources with maximum impact to create more value.
For example, I sometimes see that customers have a telephone subscription or energy contract with minimal consumption and minimal profit margins. Expensive marketing campaigns are then conducted on this. Smart if you have a well-founded expectation that these customers will bring in more turnover or margin in the coming months or years. Not smart if the expectation is that this turnover will remain minimal, and the margin remains marginal or even becomes negative. This is exactly what you are trying to achieve with CLV modelling. Mapping the current and future value of customers.
Stay critical and work with segments
When using CLV modelling in the B2B sector, be careful with the Pareto distribution (20% of your customers represent 80% of your turnover). For example, in the telephony sector. A company with 20 telephone numbers will generally have more value than a self-employed person with 1 telephone number. However, this does not mean that all self-employed people together have no value. There are also differences between customers within this group. You can solve this by working with segments and calculating the value in relation to each other within them. You can then better manage (the potential) value within the segments. This prevents you from no longer approaching a large part of your customers.