As marketing maturity has grown over time, corporations increasingly realize the value of having a 360-degree view of their customers. Traditionally, acquisition, cross-sell/up sell and retention programs have been handled by different groups within a firm.
These groups operate largely independently of one another, and the end result is often suboptimal. The organization stands to gain by focusing on prospects that will stay loyal longer and avoid the ones who are fickle or risky to serve. Similarly, it is better to focus the retention efforts and incentives on customers who are likely to buy more in the future. Therefore, organizations need to understand the lifetime value of their customers. Wikipedia says, “In marketing, customer lifetime value is the net present value of the cash flows attributed to the relationship with a customer.”
In a perfect world, an organization would want to establish a CLTV measure that accounts for the customers’ expected tenure, all likely future purchases (including cross-sell and up sell), the costs of servicing the customer (including product costs), customer service costs and losses in the event of likely default. And this CLTV measure would be available to everybody in the organization, who can then bake it into their decisions about the customer. In this scenario, even the customer service agents’ strategy for handling customer complaints/queries is driven by the CLTV measure.
As appealing as a CLTV framework might be, if not done right, it can quickly become very complex and hard to implement. The right solution depends on the data availability, analytical capabilities of the organization as well as the infrastructural constraints facing it.
Firms have varied capabilities in terms of their ability to capture data, perform analytics on the data and then make it available to their strategists and customer-facing agents. Not recognizing this could lead to a project gone haywire or a set of “cool” solutions that no one can use. For example, if customer service data is not available in the implementation platform, it might be better to use a less ambitious definition of CLTV that still adds a lot of value than to develop a measure that cannot be implemented.
Developing a CLTV framework is very important for an organization, but doing it right is equally important. Here are some best practices to follow to ensure that the CLTV framework is correctly set up.
Define actionable objectives clearly at the outset of the project
While it is really exciting to have a CLTV score to evaluate the customers and prospects, organizations should clearly lay out at the outset how they plan to use it. This is important because different firms might have varying priorities. I have seen firms with a priority of customer retention that opt to use CLTV to focus on the more valuable customers. Other organizations wish to weed out the risky and fickle customers at the time of acquisition. Determining the desired use of CLTV is important because in an imperfect world where we might not be able to do everything, the objectives can help make the right theoretical and practical compromises to get to a workable solution. For example, if the focus is acquisition of less risky customers, customer service data, though nice to have, is not as important as when one is focusing on retention.
Another reason to have well laid-out objectives is that it makes the evaluation of the program easier. For example, if the ultimate reason for putting in place a CLTV plan is to sell more to existing customers, then evaluation of the program becomes relatively less ambiguous.
This is not to say that an organization should not have multiple or complex objectives while setting up a CLTV measure, but having a clear understanding of the priorities will allow them to make better decisions and add value where it really matters.
Develop a good understanding of business, data and technology constraints
Once the objectives have been defined, it is important to understand the data and technology constraints that will limit the scope of the CLTV program. Data maturity varies by firms, and the ability to do sophisticated analysis also varies. The data constraints need to be identified, and then the objectives of the project modified if needed to meet the data constraints. The ability to perform sophisticated analysis can be outsourced if the capability does not exist in house, but the technology with which the framework will finally be implemented should be able to handle any solution developed.
The constraints and requirements of the implementation environment need to be well-understood because otherwise the CLTV score developed might not be available for use in strategic and tactical decisions. Very often, data sources are available in the development environment that seem very promising but cannot be integrated with the CRM platform. Such sources need to be avoided upfront, as they will probably show up in the solution and then mess up the implementation. Again, a model might be very good but too complex to be implemented in the CRM platform. A simple model that can be supported in the CRM environment needs to be built. Not planning for this up front will lead to rework and will have consequences for the project budget and timelines.
In addition to these data and technology-driven constraints, there might be some business limitations to what can be accomplished. A CLTV program touches many different groups of an organization, and each of those groups may have some business objectives that cannot be compromised and need to be accommodated. For example, customers in certain prespecified categories might be required to always be offered the highest price discounts, irrespective of what the analytics recommends.
A good strategy is to understand and plan for these constraints in the initial stages of the project.
Design a plan that adds the most value even though it might not be the most highly sophisticated.
It is important to chalk out a plan that is best for the organization, even though it may not be the best plan overall. It is possible that after studying the different constraints, one might come to the conclusion that a comprehensive CLTV measure would be very hard to implement, perhaps due to business strategy reasons, data unavailability or technology-related issues. In this situation, the firm could still devise a plan that takes it toward the goal of including customer value in business decisions.
For example, if customer acquisition is important and the data constraint makes it very difficult to introduce measures of future cross-sell/up sell, a modified plan that introduces some measure of future risk into the acquisition decision might be a good compromise. The important thing is to not get carried away by a grand plan that might be wonderful in a perfect scenario, but rather to build something that works and then build toward a more advanced solution.
Plan to demonstrate tangible outcomes in short to medium run
Any CLTV implementation can turn out be fairly complex because it touches so many different types of data, requires the development of a large number of complex models, involves so many different groups and then needs to integrate with the CRM platform. This invariably means a very lengthy timeline to put the solution in place. A long, drawn-out project comes with inherent risks of some stakeholders losing interest, and it is important to design it so that tangible returns can be demonstrated along the way. This keeps the different stakeholders engaged and also brings in more supporters as the results start materializing.
For example, one organization embarked on a CLTV project to primarily help with customer retention. The project took more than 18 months to complete. However, they planned for intermediate output to be in place in three months, as simpler customer retention models were first developed, then CLTV components were put in place and finally the integration with CRM was completed. So while the customer service folks did not get to practice CLTV-based retention until the very end, it started being used and incorporated in an increasing number of business decisions at a fairly early stage in the project. This not only allowed for greater organization-wide buy in as people saw the benefits of the pieces that were being put in place, but it also allowed for testing and correction based on the early usage results.
Put in a good measurement plan up front to allow for evaluation and modification
Finally, it is important to decide up front the metrics that will be used to measure the performance of the program and the criteria that will be used to determine its success. These measures should not just be based on statistical measures of analysis performance but should also include business criteria. In the customer acquisition example, it is important to verify that the statistical measures (like K-S) hold steady over time, but it is equally important to check that the stated business objective (for example, increase the number of multibuyers by 10 percent while increasing the overall base by 5 percent) also holds.
This helps us understand if things are working as planned, and if not, helps identify what needs to be remedied.
Developing a CLTV framework is an important exercise that needs a great deal of planning and commitment. It is important to spend some time upfront defining the goals and understanding the different types of constraints so that a good implementable solution is developed. A good strategy often involves starting with a simpler piece and gradually building the complexity in a planned and phased manner.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access