top of page

What is Customer Lifetime Value (CLV or LTV) and why does it matter

Updated: Sep 26, 2023

When I sat down to think about the very first piece to write for CoppettHill.com, Customer Lifetime Value was an obvious choice, as it sits at the centre of so many topics that I want to cover. In fact, most conversations about growth and marketing investment come back to the value of an individual customer or different types of customer – whether that is defining your Ideal Customer Profile (ICP), choosing how much to spend on marketing, or considering how to develop your proposition for the benefit of customers.


In simple terms though, the best use of Customer Lifetime Value in my experience is to determine what a business should rationally be prepared to ‘pay’ to acquire a customer. In today’s environment of pressure on marketing & sales budgets and an emphasis on customer retention, this feels like an even more important question to tackle, so let’s jump in.


What is Customer Lifetime Value?


Customer Lifetime Value is the profit contributed by a unique customer over their lifetime transacting with a business. Or to put it the other way round, the profit a business would lose if a unique customer had never existed. We’ll come back to what ‘lifetime’ means in practice later.


This concept has its history in database marketing - think catalogue retailers and credit card providers, those businesses where it was easiest to build a single view of customer transactional behaviour over time in days before the internet. The term ‘Customer Lifetime Value’ was used at least as early as 1988 in ‘Database Marketing’ by Merlin Stone, and first featured in the Havard Business Review in 1989.


What I really appreciate about the concept of Customer Lifetime Value is that it has stood the test of time – I recently re-read this article from 1998 (the year Google was founded) it still rings true today. This makes it one of the very few marketing or growth concepts to have made the shift from analogue to digital marketing largely unscathed. I’d argue that it has become even more relevant as marketers have become more data-driven over the past 20 years.


How to use Customer Lifetime Value?


There are four main uses of Customer Lifetime Value that I see, starting with the most frequent:

  1. To calculate ROI on marketing spend, when combined with Cost Per Acquisition (CPA) data;

  2. To compare between different customer segments (which can tell you either attributes of customers that make them more attractive to your business and/or groups for whom your proposition is a better fit);

  3. To measure the impact of historical business changes over time – seeing how Customer Lifetime Value changed; and

  4. To model the potential impact of future business changes – to combine different assumptions and forecast customer profitability scenarios.

Whilst LTV is a great concept to embed in both daily decision making and big strategic decisions, I don’t think it is well suited to routine monthly Management/Board reporting. As a lagging, historical measure it is unlikely to move by much month to month, so it is better suited for an annual strategy day, or to be operationalised into marketing decisions e.g. Paid Search bidding for different customer segments or partnership commercial models.


How to calculate Customer Lifetime Value?


To calculate Customer Lifetime Value, you need to consider all revenues associated with a unique customer, then remove all direct costs, a fair share of variable operating costs, and any reacquisition costs for subsequent transactions.


customer lifetime value formula

The specifics here will vary by business model and for each customer, but to give some more examples:


Revenue – this should include both the main transactional revenue from a customer, but also any ancilliaries or one-time income, for example cancellation insurance added to a holiday booking, or one-off implementation fees associated with a SaaS subscription. Don’t forget to also allow for discounts offered to customers – only count the true revenue received.


Direct costs – the best way to think about this is your gross margin – either the costs of physical goods or services, as well as staff costs allocated to a specific transaction. Don’t forget to also include the costs associated with ancillaries or one-time income, as well as things like bank fees/payment processing, logistics, insurance, returns etc.


Fair share of variable operating costs these are costs that you might not allocate to specific customers on a day-to-day basis, but which broadly correlate to the number of customers you are serving – for example Customer Service or Support teams. This is the one where there is normally the most debate about what to include in an LTV analysis.


Reacquisition costs – some repeat transactions will have additional marketing or sales costs, for example the staff cost to secure a renewal in a SaaS business, or a price comparison website commission fee for an insurer.


Getting hold of the data put this analysis together takes time, in my experience it is normally easier in B2C than B2B businesses as you will typically already have access to customer-level revenue data. You may have to be creative - I’ve had to use invoice level data from finance systems or stitch customer data together from multiple sources - but I've always managed to find the right information in the end.


Some of the inputs into a Customer Lifetime Value calculation will be at unique customer level (normally revenue data), for others you will need to make assumptions for segments of customers or for everyone (normally cost data).


There are many tools available that claim to have some version of Customer Lifetime Value analysis available ‘out of the box’, but I prefer to start by calculating it directly. There will always be limits to analytical capabilities with a set of pre-configured reports/dashboards, and most will make at least one of the common mistakes I talk about later on.


What does ‘Lifetime’ really mean?


Every customer’s lifetime with your business will be different – and just because they may have stopped transacting with you for now, doesn’t mean they will never come back. To get round this dilemma, I use the concept of a ‘lifetime window’ in my Customer Lifetime Value analysis. This is a standard period of time, often 3 or 5 years, from the first transaction with a customer. It allows for standardised analysis and comparison between unique customers or customer segments.


Determining which time period to use for the ‘lifetime window’ isn’t an exact science, but is a trade-off between the length of the window and how many customers will be eligible for the analysis. If we set a 5 year ‘lifetime window’, our historical analysis won’t include customers acquired less than 5 years ago. You should only decide this once you’ve assembled your historical data – and is why should always build the longest time-series of data as possible, within reason.


customer lifetime value window

This makes historical Customer Lifetime Value analysis particularly challenging for new businesses. In these situations I’ve used a much shorter window, sometimes 12 months or less.


You may have seen examples of Customer Lifetime Value analysis which use a method of dividing annual revenue by an expected annual churn rate, sometimes also with a discount factor. Whilst this often produces very high estimates of Customer Lifetime Value (great when talking to potential investors), I’d always stick to using actual, historical behaviour if you are trying to make strategic choices.


The pattern of revenue will vary based on your business model, or potentially within your business – is your product/service an annual purchase, a frequent purchase or a subscription? Some businesses may even only transact once with the vast majority of customers (think divorce lawyers or funeral directors!). Using a ‘lifetime window’ will help to standardise any analysis.


What is a good Customer Lifetime Value?


The answer is clearly ‘it depends’. This will entirely depend on your business, and I wouldn’t advocate using Customer Lifetime Value as a benchmark metric in isolation.


There are some obvious rules of thumb however – within a niche of comparable propositions, you will see higher lifetime value for those businesses with (i) better margins, (ii) better repeat rates, and (iii) better ability to up-sell/cross-sell to customers.


What segments should I consider when analysing Customer Lifetime Value?


One of the most powerful questions you can answer with Customer Lifetime Value analysis is “Who are our most valuable customers?”.


To answer this, you can analyse the relative LTV of different segments based on different customer-level dimensions. These could be ‘attributes’ such as age, location or industry vertical; or ‘behavioural’ such as what the customer purchased first or which marketing channel they came from.


This is a process of elimination, test many different dimensions and narrow down to the ones that make a difference. When you find the combination of dimensions that allows you to create a segmentation that balances the best spread of LTV vs equal distribution of customers, you can start to operationalise this. This could be with just one characteristic, for example risk type in an insurance business, or a combination of 2 or more dimensions. It is best to not over-complicate your segmentation at first as it will be harder for your stakeholders to understand and then hard to operationalise.


Make sure that you always pay attention to any outliers in your analysis - very low or loss making customers, or super profitable customers. These can lead you either to great insights or bugs in your analysis that need fixing, and sometimes both!


What are the common mistakes with using Customer Lifetime Value?


This isn’t an exhaustive list, but there here are five of the most common mistakes I’ve seen when reviewing Customer Lifetime Value metrics:


  1. Only considering revenue – the most common mistake, where LTV is stated at revenue rather than profit level. This can lead to poor decision making and ultimately value erosion.

  2. Ignoring reacquisition costs – repeat purchases from customers will often carry additional costs, sometimes very significant ones, for example in businesses which spend significantly on advertising. Did the customer repeat purchase because they were loyal or because they saw your advert again when searching generically online?

  3. Not factoring in customer service costs – in some business models, there is a significant amount of staff cost required to service existing customers, which can be ignored in LTV analysis. There is some subjectivity on where to draw the line, but a share of the cost of large teams such as Customer Service or Support should be factored into your analysis

  4. Ignoring changes in a business over the historical period e.g. the introduction of new revenue streams or a major change in pricing. This can complicate analysis, but if you want to use this metric to make choices today about the future, you should make adjustments to historical data to best reflect the future value of customers you have acquired today. In practice that might mean re-stating historical revenue for some customers.

  5. Using Customer Lifetime Value to make decisions in isolation. You aren’t seeing the whole picture if you do this – for example you could have a great picture on “Who are our most valuable customers”, but they could represent only a tiny proportion of your potentially addressable market, or have a very low conversion rate vs other customer segments. Make sure to combine LTV analysis with market size and conversion rate data.


How to increase Customer Lifetime Value?


I’ve written a separate piece about this which you can find here.


If you’d like to discuss how you can better understand and use Customer Lifetime Value in your business, please Contact Me.


All views expressed in this post are the author's own and should not be relied upon for any reason. Clearly.

181 views

Comments


Subscribe to get my ideas as they are published

Thanks for subscribing - I normally publish ideas at the start of each week.

bottom of page