Customer Lifetime Value
The basic idea of CLV is that if the marketer understands what it costs to acquire, maintain, and service a customer, he or she can make a reasoned decison about how much to spend to market that customer. Data must be collected for as long as 3 years in order to begin developing CLV and marketing programs based on it. The cost of communicating with the customer determines marketing costs. The contribution margin from each customer is determined, and the amount of communications with the customer across all channels is taken into account. The profit from the customer is based in the frequency of purchasing at a particular contribution margin (revenue times contribution percent) less the costs of marketing to the customer. The discount rate is applied and the net present value of a future stream of profits- CLV- is computed. Then the marketer determines actions that can be taken to increase CLV.
The Testing Process
- Reasons for conducting a test
- Design a test
- Establish Test Metrics
- Execute and monitor the test
- Analyze and report test results
- Make marketing decisions
Data can be derived from: purchases, census, associations, research firms, purchase data, survey/questionnaires, focus groups, interview, and websites.
There are actually numerous databases that should all be linked through some type of a central repository, such as a data warehouse, so that information can be provided on demand to decision makers.
The data warehouse can be conceptualized as having 4 basic components.
- The marketing management databases drive marketing programs, such as customer database, prospect database, retailer/distributor database and promotions database.
- The marketing support databases provide additional data that are important for decision making purposes, such as competitive intelligence database, marketing research database.
- Other areas of the business contribute databases that are crucial to the functioning of marketing programs and customer service, including the order/transaction database, inventory database, sales force management database, and project/personnel database.
- Externally purchased or linked databases provide additional valuable data, such as universal product code/scanner database, geodemographic or other syndicated databases, and third party databases.
Data Mining is used to produce customer information not previously known, or perhaps even previously hypothesized from large databases. Data mining produces new knowledge because it looks for patterns in data that might not be revealed by traditional statistical analysis. Techniques like decision trees and neural networks may be used, as well as techniques commonly used in marketing research, such as regression and cluster analysis. Marketing managers do no have to deal directly with the complex analytics.
Each step requires data, and more specific data in the process in order to reach customization stage.