• June 7, 2022
  • By Lee Barnes, leader, Data Insights team, Paytronix

Know Your Customers’ Lifetime Value

Article Featured Image

The metric of customer lifetime value (CLV) has been put to great use by major brands like Disney and Amazon. With a deep well of first-party data and a team of data scientists, they can predict how much each customer is likely to spend in the future. That helps these brands make informed decisions about customer acquisition and retention, as well as measure the results of each program.

At its core, CLV offers a way to understand customers as individuals rather than as part of broad groups. A customer who comes in twice a week for coffee is fundamentally different than one who comes in once every two weeks for a full breakfast. When done correctly using the right data, CLV provides insights on how much a customer is worth to a particular brand. If you understand the CLV of each customer, you effectively have a list telling you which customers offer the most promise and where to invest your marketing dollars.

Calculating CLV

CLV requires good first-party data, and that’s just what loyalty programs provide. This data reveals exactly how much loyalty members are spending, how often they’re visiting, and the length of time between each visit. Put more simply, it spotlights the core concepts of CLV: recency, frequency, latency, and spend.

Paytronix has leveraged its 20 years of data and experience in the loyalty space to devise a method of calculating CLV that brings this powerful metric to restaurants and convenience stores. It starts by developing a baseline for the brand itself, and then through using AI to drive individual action, the metric enables deeper understanding of what this means for each individual customer.

The objective isn’t creating a metric for the industry or even for a specific group of restaurant concepts. This is a bespoke metric that focuses on each brand. To arrive at it, the Paytronix data insights team ingests years’ worth of data from a given brand and runs it through a machine-learning algorithm to establish the CLV baseline. This basic scaffolding provides context for value, accounting for whether it’s a fine-dining restaurant that sees guests spend hundreds of dollars on an occasional visit or a coffee shop that people frequent several times a week but spend only $5 each time.

Getting Started with CLV

The Paytronix model consists of four basic inputs:

Once we have the brand metric set, we can use the same four core concepts of recency, frequency, latency, and spend to determine the value of a guest in the future for any given time period.

A guest who came in yesterday, spends 5 percent below average, and usually visits two or three times a week has a different CLV than one who traditionally spends 20 percent above average but hasn’t ordered in a month. The loyalty programs used to drive the behavior of these customers will be different, and the spend for each program will vary as well.

Take a hypothetical guest who comes into a store regularly and spends about $15 each visit. In the chart below, the yellow line tracks the guest’s lifetime spend to date, while the blue line represents the guest’s future value. You can see that the CLV increases when the guest establishes a regular visit cadence and then declines as latency rises, reflecting the increased likelihood that the guest has lapsed.

Of course, the more data we have about a consumer, the more accurate we can be regarding the CLV.

When testing the model, the data insights team focused on a single brand and began with data from 2015 to train it. The predictions were then measured against what actually happened in each subsequent year. We found that the predictive models ran very close to reality in 2016, 2017, 2018, and 2019, but there was misalignment in 2020. Why? Well, the changes brought on by the pandemic hadn’t been built into our model. Predictions in 2021 returned to being reliable, and we continue to limit the use of 2020 data for training models.

How CLV Can Be Used

Once we have a baseline, we can use CLV in a number of ways to achieve better programs.

Identify Top Customers: CLV generates a list of your customers and their value, from most to least valuable. This enables you to clearly identify those who are of most value to the brand and reward them accordingly.

Segment Effectively: Most types of segmentation group people into different bands, whether it be by type of spending, visit frequency, or purchasing category. CLV allows for more accurate segmentation to create more effective programs.

Optimize Acquisition: Acquisition strategies generally cast a wide net to bring in a large group of customers and encourage them to join a rewards program. CLV enables a specialized approach in which marketers focus on those customers who offer the most long-term value.

Realize Lift: CLV serves as a simple metric for optimizing campaigns by focusing on long-term value. It’s one thing to run a program that increases visit frequency, but understanding the ongoing impacts of that change will shape the program’s approach.

Knowledge of customer lifetime value has become an integral part of the marketer’s toolbox. By having good predictive information about how much a customer is likely to spend over a given period of time, a brand can make better decisions regarding the cost of its retention programs.

We often hear marketers complain that their programs have been cost-capped by the finance department because of the perceived impact they deliver. With CLV, the quantifiable long-term results should make any CFO smile.

Data analytics expert Lee Barnes leads the data insights team at Paytronix, a leading provider of reward program solutions whose guest engagement platform helps more than 300 restaurant and retail chains manage and grow more than $18 billion in guest spend. Barnes is a self-confessed data geek that can often be found digging into the data with his team to optimize guest engagement with more than 165 million loyal guests—through mobile, social and today’s most innovative digital marketing tools. His mathematics degree and his MBA from Harvard Business School gives him the unusual ability to both execute complex analyses and translate the results into ideas that business leaders can use.

CRM Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues