Most companies understand that retaining customers through difficult economic times is extremely important. In most industries customer attrition can be anywhere between 10% and 40% in any single year. Mobile phone operators for example typically report numbers between 30 and 35% per year for mobile phone subscriptions, with some groups of customers even reaching levels higher than 50%. This not only results in a direct loss to the business, but also in a loss of any potential future income these customers may have brought to the company. What is more, attracting new customers is typically 5 to 10 times more expensive than keeping existing customers. This means that understanding why some customers leave the business and what we can do to retain them is extremely important and has a direct impact on the company's revenue.

Survival analysis is a technique introduced in the biomedical sciences to estimate survival probabilities over time and to understand why survival rates differ between treatment groups. Increasingly this technique is being used in customer relationship management to understand who is most likely to attrite but also, very crucially, when this is most likely to happen. Having this information means that we can target the customers that may be thinking about cancelling their relationship with the company at the right time. But perhaps more importantly, not having this information means that you may loose some of the customers that may have brought value to you in the long run.

Regression techniques have been adapted to survival analysis to uncover which customer characteristics drive survival rates. Factors such as socio-economic profile, customer attitudes, subscription history or past contacts with the company can all be essential in predicting how much more or less likely a customer is to stay on when compared to his peers. Understanding these drivers can also help us to tailor the offers we make the customer in the most efficient way possible, to make sure we try and retain the customer at risk.

Customer lifetime value analysis (LTV) combines survival probabilities with the average spend of a customer throughout his lifecycle, and helps us to estimate what the total value of a customer is to the business. Typically, loyal customers will spend more over time, and even go on to purchase other products in your range. LTV analysis can help us to understand what the true costs and returns of a direct marketing campaign are. Given the response rates and cost of a campaign, survival analysis will tell us how many of the targeted customers will remain over time and LTV will tell us what the net value of these customers is. This means that we can then accurately estimate the breakeven point and ROI for any campaign. Moreover, survival analysis and regression can help us to find the correct mix of customers to target and minimise the breakeven point and maximise ROI.

 In summary, customer lifetime value, combined with survival analysis will help you to:

• Estimate the ROI and break-even point for a marketing campaign
• Find the correct mix of customers to target in a campaign
• Know when to target particular customers and which offers are most likely to help retain them
• Make optimal use of the information stored in your CRM 
• Stay one step ahead of the competition by retaining your most valuable customers and attracting the right mix of new customers
• Minimise the loss to the business through attrition and maximise customer loyalty and profit