The basic idea of Customer Lifetime Value (often abbreviated as CLV or LTV) is easy enough to grasp: decisions about acquiring and retaining customers should be based on the value generated over the entire duration of the customer relationship.
But once again, the devil is in the detail. There are various alternative methods for calculating the CLV, ranging from overly simplistic to downright painstaking, and many companies are not certain which method to choose. CLV is also sometimes hailed as an oracle that can answer all customer-related questions, but companies are often unclear about which specific business problems they are trying to tackle with CLV. It’s no wonder then that while most companies feel CLV is important, only few have succeeded in using it effectively(1).
The path to using CLV to drive business benefits successfully can be broken down into the following five major steps.
Step 1: Define the objective.
As is often the case, the path to success starts with asking the right questions. The key question here is what business benefits are you hoping to drive with CLV? Are you trying to optimize your marketing mix? Differentiate customer service levels? Or perhaps nudge customers toward behavior that fosters loyalty? A clearly defined business objective is the beacon that will guide you through the rest of the process.
Step 2: Choose your CLV metric.
Once you know which business benefits you are hoping to drive, you can design your CLV metric accordingly. You will need to consider factors such as: Should you measure revenue or profit? Should you look at past facts, or try to predict the future? And should you take into account the influence that one customer can have on other prospective and existing customers?
Step 3: Build your CLV metric.
Now it’s time to roll up your sleeves get your hands dirty with data. The size and difficulty of the job depends on the CLV metric you have chosen. Fortunately, the first CLV calculations can often be completed as a manual quick-and-dirty exercise, and can be automated later once the business benefits have been proven.
In any case, sooner or later you will probably run into problems with data: some data is not available, or the data quality is not optimal, or recent data is not comparable with historical data etc. It is important not to get bogged down with trying to obtain perfect input data, because there’s no such thing. This is particularly true for CLV, which inherently demands input data over long periods (often several years). Be prepared to make simplifications, estimations and compromises, and keep pressing forward. The trick is matching the data you have to the question you are trying to answer.
Step 4: Validate the business benefits.
Once you have invested the time and effort into calculating you first CLV values, it is time to put them to the test. You must demonstrate that using CLV can generate the kind of business benefits you set out to achieve. If your objective was to optimize the marketing mix, did you find meaningful CLV-differences between marketing campaigns, and how much can you gain by spending your marketing budget more wisely? If your objective was to differentiate customer service levels, is the premium service generating more repeat business from the high-CLV-customers? And if your objective was to nudge customers toward loyalty-fostering behavior, do your interventions indeed reduce churn and/or increase repeat purchases? Whenever possible, consider using A/B-testing for validating the business benefits.
Step 5: Automate, improve and expand.
Since you have now proven that CLV can generate tangible benefits for your business, you can go full monty with the implementation. You can automate the CLV calculation process, and integrate the CLV metric into your business processes. This is usually not just a technical exercise, but also an organizational and cultural challenge. For example, you might need to redefine the incentives for your sales reps and/or customer care agents.
As time goes on, you can assess the accuracy of your earlier CLV predictions, and tweak your models accordingly. You can also experiment with other use-cases for utilizing your CLV metric, as long as they are compatible with the way your metric has been constructed.
Stay tuned, there’s more to come!
This is the first in a series of blog posts about harnessing the power of CLV. In the next post, we will dive deeper into the various kinds of business benefits that companies can hope to derive from CLV.
(1) According to research by Criteo, 98% of companies believe that CLV is important to their marketing strategy, and 93% are trying to measure it, but only 24% have succeeded in effectively monitoring CLV.