Using Customer Lifetime Value to Transform not Wreck your Business

wrecking-ball-wall

When calculated correctly, Customer Lifetime Value is possibly the most commercially valuable insight at the disposal of the marketing team. However, if calculated wrongly it can be a fast-track way to wreck marketing effectiveness and destroy a business!

An accurate view of Lifetime Value can be used as a primary measurement to predict an individual's potential value to the business and to track the success or failure at each step of the customer's journey in obtaining that potential value.

Why is Customer Lifetime Value (CLTV) so Valuable?

Rather than look at simple campaign metrics in isolation, such as clicks, opens and conversions, which really only provide you with a snapshot of a specific campaign, a dynamic CLTV calculation can be used to track the success of the entire marketing function in a holistic way, including all acquisition, cross-sell, up-sell, engagement and retention initiatives.

For organizations that get CLTV right, it enables them to pivot everything they do around the CLTV model. For example:

  • If you improve retention then your average CLTV increases.
  • If you reduce your acquisition costs or cost to serve then your CLTV also increases.
  • If you improve your cross-sell results your CLTV will increase again.

All of these initiatives can be captured and analyzed using a dynamic CLTV model. Therefore understanding how Lifetime Value is effected by channels, advertising and campaigns will help to drive reduced acquisition costs through the business.

Moreover, it can be used to recognize customers who are potentially worth more to the business than others, and make decisions about how to segment and treat those high-worth customers differently to those who don't.

How do you Calculate Customer Lifetime Value?

A basic formula for calculating Customer Lifetime Value is:

Annual profit contribution from a customer MULTIPLIED BY 
(Number of years they remain a customer MINUS 
The initial cost of acquisition and service)

What information is needed to calculate this?

  1. Initial cost of acquiring the customer
  2. Profit contribution of the customer
  3. Cost to service the customer
  4. The customer's retention rate

This formula probably over simplifies things. For example, understanding the 'Initial Cost of Acquiring the Customer' relies on being able to attribute the correct costs. Customer Attribution is easily said than done, particularly when you are marketing across many disconnected channels. Attempting to determine if it was the direct mail, email or digital advertising campaign that acquired the customer can be difficult and may require assumptions, percentage weightings for first and last touch, time assumptions etc.

Other problems with calculating lifetime value become apparent too when you start to query the underlying data. For example, if you want to use your calculations to predict the lifetime value of NETT new customers you will be in for a shock if you do not factor in someone's age. An 80 year old new customer is clearly not going to have the same potential value over time that a millennial would have.

Gender, age group, region, income, etc can all be factors that can impact the validity of your model.

Here is an example for calculating Lifetime Value, which although old, I particularly like and is available online*:

Customer Lifetime Value - formula example

  • pt = price paid by a consumer at time t
  • ct = direct cost of servicing the customer at time t
  • i = discount rate or cost of capital for the firm
  • rt = probability of customer repeat buying or being “alive” at time t
  • AC = acquisition cost
  • T = time horizon for estimating CLV (Customer Lifetime Value)

Download our 'A Guide to Predictive Analytics' eBook

From the importance of good data as a foundation for your strategies, through to building your own predictive models, this 'A Guide to Predictive Analytics' eBook looks at how a range of applications can both reinforce customer loyalty and increase the probability of purchase.

Download now

A Guide to Predictive Analytics

The source for this formula is a fascinating article that delves into many ways for calculating CLTV, and how to combine RFV models with CLTV. I would recommend taking a look, as it starts to explore many of the variables and factors that you may need to include. Underlying this model is the following set of assumptions:

  • A customer’s relationship with the firm has two phases: He or she is “alive” for an unobserved period of time, and then becomes permanently inactive.
  • While “alive,” the number of transactions made by a customer can be characterized by a Poisson process.
  • Heterogeneity in the transaction rate across customers follows a gamma distribution.
  • Each customer’s unobserved “lifetime” is distributed exponential.
  • Heterogeneity in dropout rates across customers follows a gamma distribution.
  • The transaction rates and the dropout rates vary independently across customers.

* Source:
Modeling Customer Lifetime Value, Journal of Service Research, Vol 9, No. 2, November 2006, 139-155

Different Ways to Calculate CLTV

The example above requires a lot of assumptions and therefore wanted to find out how organizations are practically putting the theory of Lifetime Value into action. Speaking to an experienced database marketing professional that has utilized BlueVenn's analytics and marketing automation technology at a number of leading US media organizations, we spent time discussing CLTV at length, and how he has used CLTV practically within those businesses. 

The first thing he pointed out is that he uses three different equations to calculate CLTV and then aggregates the values to create an average. In his view, this is essential due to the fact that there are different ways to calculate CLTV, which give you different results, and as he put it to me - “Who knows which one is right?!”

Thankfully he shared these calculations with me for this blog and, although I've not given you all the inputs for the formula, it is an example of how one customer put through three different CLTV calculations can provide three very different results - $1,395.59, $16,306.87 and $4,405.09.

Customer Lifetime Value - formula and equations

By averaging these three answers therefore he has found a way to use the aggregated CLTV to analyze the entire customer database and, in time, make predictions which have had very positive effects on revenue, and is the driving force behind future decisions for acquisition, segmentation, channels and retention.

Putting Customer Lifetime Value into Practice

Here are some examples that were used at one US media organization to analyze age groups and payment methods through the CLTV model:

#1 - Analyzing age of customer shows a significant difference in AVG Lifetime Value

LTV Triggers for branching - age based

The first analysis found a $1,400+ difference in the average CLTV between Younger and Older age segments across their database for a 3 year time period.

This informed the business of the need to appeal to their younger audience with a different message and channel, and resulted in a year-long branding and content exercise to build the database of younger subscribers at the start of their Potential LifeTime Value (PLTV).

#2 - Payment method also showed a significant difference in AVG CLTV

LTV-Triggers-for-branching-payment-based

The second piece of analysis was that the average CLTV in a sample data of EZPay customers was almost $2,000 more than that of a pay-by-mail customer.

As a result, the organization gave unprecedented focus to accelerate customers, across all segments, towards using EZPay through incentives and timely offers that accelerated the transition in payment method.

 

Some Tips Before you Rush into the CLTV Game

I was impressed by the simplicity of these two examples of analysis, but as my colleague rightly pointed out, it has taken a long time to test and perfect the CLTV calculations to get to this point. I therefore asked him to provide his tips for this blog, and here is some sound advice directly from someone that has implemented it successfully:

  • "Introducing LTV into your campaigns can actually do a lot of damage, but when it’s done right, it will be an incredible tool."
  • "Begin with the assumption that marketing actually works, and therefore the customers with the greatest Potential LifeTime Value (PLTV) - not your current high value customers - need your focus."
  • "LTV helps address message changes to handle issues such as “we lost this high value customer, how much can we afford to spend to win them back?”, or “which customers should I be targeting to maximize ROI?”
  • "In addition to these questions, correctly implemented LTV models can bring logic and evidence to marketing decisions, as well as elegance to the strategy of who to target, why, when and with what offer."

So I revert back to the words at the start of this article: When calculated correctly, Customer Lifetime Value is possibly the most commercially valuable insight at your disposal. However, if calculated wrongly it can be a fast-track way to wreck marketing effectiveness and destroy a business!"

[Cliché time] It's all about the Data!

I hope you’ve enjoyed this blog and it has got you thinking about Lifetime Value, and more importantly how to inject this into your own marketing strategy. However, like most things in marketing, it all starts with having trustworthy data at your disposal, and the capabilities and tools to analyze your customer data effectively. Without access to every byte of data in a Single Customer View you could potentially be in the camp of marketers that can potentially “destroy a business.”


Putting your Customer Data to Work with Predictive Analytics Webinar

On-demand Recording

Predictive analytics can be a great way to uncover historical trends in your customer’s transactions and behaviors to help better predict what they will do or buy next.

This webinar shows examples of customer segmentation using predictive models and how to use these to improve the success of your marketing campaigns.

Watch now

customer-data-platform-data-quality
Topics: lifetime value customer retention customer loyalty customer lifecycle