Is customer lifetime value an outmoded metric?

New year, same old CRM metrics, right? Reflecting on the reverence the marketing community has given to Customer Lifetime Value (CLV), it may be time to ask whether this  metric is truly serving marketers (and, by extension, customers). 

Of course, CLV has its benefits: it’s clearly been (and will continue to be) an important visualisation for businesses in the future. The method is simple to work with, it sums up customers’ value to date, it easily segments clients into high and low value and, foremost, plays a key role in businesses’ reporting and customer management armoury. Besides, it has been a crucial KPI for many companies’ retention strategies by serving as a determining factor for promotional spend. 

The metric will surely continue to be a comprehensive measurement to commonise business conversations, and marketers will continue to encounter it as they jump from one company or sector to another. But here’s the problem with relying solely on traditional CLV: it's retrospective, and just because a customer was valuable in the past doesn’t mean they will be in the future. 

While a vital component in the marketing playbook, it’s difficult for marketers to influence CLV with such a disconnect between strategy and execution. Whatsmore, CLV focuses more on ‘baskets’ than it does on behaviours, which is a missed opportunity to increase customer loyalty long-term to say the least. 

I believe that’s the aspect that desperately needs a reboot, and here’s why data capability and analysis should come in to support. With machine learning, marketing teams can go beyond CLV’s ‘business as usual’ and create forward-looking and highly operationalised strategies. By analysing full datasets and looking at customers’ past data, models can predict what users will do in the future, and allow marketers to focus on identifying triggers and marketing tactics to modify behaviour. 

This results in turning a measurement that previously categorised customers based on historical data into actionable insights to increase individual customers’ future value. Besides, with an increasing number of brands already investing substantial amounts into toolkits and infrastructures that capture behavioural first-party data, this new approach to CRM means further leveraging data and folding it into a forward-looking value segmentation. 

At Planning-inc, we have implemented this solution for multiple retail brands, and the reboot looks something like this: we first try to understand first-party data availability, and once this has been settled, we define what ‘value’ means to the business - whether that be revenue, engagement, subscription tenure or something else. By doing this, it becomes possible to identify key behaviours that can be used as signals of intent, which is no easy task as indications might vary between customer-level site engagement, marketing engagement, or customer service information and more. 

And then machine learning comes into play – analysing the correlations between behaviour and value, and coming up with a value score. Each customer is then plotted on a value trajectory, with the model creating a targeting framework to manage clients based on their expected future value -  making it simple for the marketing teams to assign promotional spend and achieve the highest ROI. 

Crucially, the solution then identifies a series of Next Best Actions (NBAs) you need to put in front of each customer that are most likely to increase their value. These NBAs are outputted as personalisation and targeting attributes that are integrated with a campaign’s management platform in real time, driving messaging that works for both the customer and the business.

To successfully implement this method, businesses need usable datasets and careful campaign planning, which obviously requires a certain level of both data science and marketing planning expertise to ensure it is operational and effective. It’s also worth noting that Future Value Modelling works best for companies that have access to behavioural data, as well as products or propositions that have a degree of engagement beyond linear transactions. 

The rewards are groundbreaking: we have seen the impact this approach has on managing customers across their whole lifecycle, driving £100ks in incremental revenue for brands. We even saw one gaming brand increase first to second play retention by 44% by focusing on individualised NBAs, targeting customers with messaging in the onboarding program, and ultimately creating sticky and engaged customers in the company’s early life.  

So what does this mean? Well, that marketers can now finally manage customers in a way that better connects strategic and reporting requirements with marketing execution. Marketing teams can now enjoy the benefits of identifying and activating the behaviours that will increase value over a specific time period rather than basing this on historical behaviours.