Customer Lifetime Value Is the Wrong Number to Optimise For

Customer lifetime value measures how much revenue a customer generates over their relationship with your business. Most companies calculate it, fewer use it well, and almost none ask whether it’s actually the right number to be chasing. CLV is a useful lens, but treating it as a growth target in its own right tends to produce a particular kind of marketing: expensive retention programmes, loyalty mechanics, and re-engagement campaigns that prop up relationships that probably shouldn’t exist in the first place.

The more useful question isn’t “how do we maximise CLV?” It’s “what kind of customers do we want, and what does it actually cost us to serve them well?” That reframe changes everything about how you approach acquisition, retention, and growth strategy.

What Customer Lifetime Value Actually Measures

CLV, sometimes written as LTV or CLTV, is an estimate of the total net value a customer delivers over the entire time they do business with you. In its simplest form, it multiplies average order value by purchase frequency by customer lifespan. More sophisticated versions layer in gross margin, discount rates, and probabilistic churn modelling.

The formula isn’t the problem. The problem is what people do with the output.

I’ve sat in enough planning sessions to know the pattern. Someone runs the CLV numbers, identifies the top decile of customers by predicted value, and the room immediately pivots to “how do we get more of these people?” It sounds rigorous. It isn’t. What you’re usually looking at is a historical average dressed up as a forward-looking prediction, calculated on revenue rather than margin, and based on cohort behaviour that may not repeat in a different competitive environment.

That’s before you get to the cost-to-serve problem, which most CLV models ignore entirely. A customer who spends £5,000 a year but calls your support team every fortnight, returns 40% of what they buy, and negotiates hard on every renewal isn’t worth the same as a customer who spends £3,500 and never contacts you. The CLV model will tell you the first customer is more valuable. Your operations team will tell you a different story.

Why CLV Models Break in Practice

When I was running agencies, we used to build CLV projections for clients as part of go-to-market planning. The models were always more confident than they should have been. We’d present a range of scenarios, clients would anchor on the middle number, and that number would end up in a board deck as a target. The problem is that CLV is a retrospective measure applied prospectively. You’re using what past customers did to predict what future customers will do, in conditions that are already different.

Three things tend to break CLV models in practice.

First, churn is non-linear. Most models assume a relatively stable churn rate across the customer base. In reality, churn clusters. It spikes in the first 90 days. It spikes again at renewal. It spikes when a competitor launches something new or when your product has a bad quarter. A flat churn assumption produces CLV estimates that look reasonable in aggregate but are wrong at the individual customer level, which is where the decisions actually get made.

Second, acquisition channel affects lifetime value more than most people account for. Customers acquired through paid search behave differently from customers acquired through referral, which behave differently from customers acquired through content or brand. If you blend all acquisition channels into a single CLV calculation, you’re averaging out a signal that would actually tell you something useful about where to spend your acquisition budget.

Third, and most importantly, CLV models almost never account for the cost of the marketing and retention activity required to achieve that lifetime value. A customer who stays for five years because you’ve run a loyalty programme, sent them weekly emails, and offered them a discount at every renewal point isn’t generating the same margin as a customer who stays for five years because your product is genuinely excellent. The CLV number looks identical. The business reality is completely different.

If you’re thinking about how CLV fits into a broader commercial growth framework, the Go-To-Market and Growth Strategy hub covers the underlying principles that connect acquisition economics, retention, and long-term revenue planning.

The Acquisition Trap: When CLV Justifies Bad Spending

One of the more dangerous uses of CLV is as justification for high customer acquisition costs. The logic runs like this: if a customer is worth £10,000 over their lifetime, we can afford to spend £2,000 to acquire them. A 5:1 LTV:CAC ratio looks healthy on paper. But that calculation assumes the CLV estimate is accurate, the margin assumptions are correct, the churn rate holds, and the retention activity required to achieve that lifetime value has been properly costed. Change any one of those assumptions and the maths falls apart.

I’ve seen this play out in subscription businesses particularly. A company acquires customers aggressively on the basis of projected CLV, the cohort underperforms expectations, and suddenly the payback period stretches from 12 months to 30 months. At that point, the business isn’t growing, it’s borrowing from the future. The CLV model didn’t cause the problem, but it provided the intellectual cover for the spending that created it.

The more honest version of acquisition planning separates what you know from what you’re assuming. You know your current CAC. You know your gross margin. You know your churn rate for the first 12 months, because you have actual data on that. Everything beyond 12 months is a projection, and projections should be treated as scenarios, not targets. Tools like those covered in Semrush’s breakdown of growth strategy approaches can help stress-test acquisition assumptions before you commit budget to them.

What High-Value Customers Actually Have in Common

When you strip away the CLV formula and look at what actually differentiates high-value customers from average ones, the answer is almost always the same: fit. The customers who stay longest, spend most, and cost least to serve are the ones for whom your product or service genuinely solves a real problem. They’re not staying because of a loyalty programme. They’re staying because leaving would create a problem for them.

That sounds obvious, but the implications are significant. It means the most important thing you can do for long-term CLV isn’t to build better retention mechanics. It’s to get acquisition right in the first place. Acquiring customers who are a genuine fit for your product will always outperform acquiring a larger number of customers and then trying to retain the ones who aren’t.

This is where I think a lot of performance marketing goes wrong. There’s a tendency to optimise for volume at the top of the funnel and then treat retention as a separate problem downstream. But if you’re acquiring the wrong customers at scale, you’re not building a retention problem. You’re building a structural margin problem that no amount of email automation will fix.

The businesses I’ve seen grow sustainably, across B2B and B2C, across product and service categories, share one characteristic: they are genuinely good at what they do, and their customers know it. If a company truly delivered on its promise at every interaction, that alone would drive repeat purchase and referral at a rate that most retention programmes can’t match. Marketing is often used as a blunt instrument to compensate for product or service gaps that the business hasn’t fixed. CLV models can obscure this by making retention activity look like value creation when it’s actually cost absorption.

The Metrics That Matter More Than CLV

None of this means CLV is useless. It’s a valuable input into planning, particularly for understanding the upper bound of what you can afford to spend on acquisition and for comparing the relative value of different customer segments. But it works better as a diagnostic than as a target.

The metrics that tend to give a more accurate picture of customer health are more granular and less flattering.

Margin per customer, not revenue per customer. This is the number that actually tells you whether a customer relationship is worth maintaining. Revenue is a vanity metric at the individual customer level if you’re not accounting for the cost of goods, support, returns, and retention activity associated with that customer.

Net Revenue Retention, particularly in subscription or recurring revenue models. NRR measures whether your existing customer base is growing in value over time, accounting for expansion, contraction, and churn. An NRR above 100% means your existing customers are generating more revenue this period than last, even accounting for losses. That’s a more direct measure of customer health than a projected CLV figure.

Time to second purchase, in transactional models. The gap between first and second purchase is one of the strongest predictors of long-term retention. Customers who buy twice quickly are significantly more likely to become long-term customers than those who don’t. This is a metric you can act on in the short term, unlike CLV which is inherently long-horizon.

Referral rate, which is the metric that most CLV models don’t capture at all. A customer who refers two new customers is worth materially more than their own CLV suggests. Businesses that track referral behaviour by customer segment often find that their highest CLV customers are not their most valuable referrers. The customer who refers enthusiastically is often a mid-tier spender who genuinely loves the product, not the high-volume purchaser who treats you as a commodity supplier.

Approaches to scaling customer intelligence and growth measurement are covered in depth across the growth strategy resources on The Marketing Juice, including how to build measurement frameworks that connect acquisition and retention without conflating them.

How to Use CLV Without Being Misled by It

If you want CLV to be useful rather than decorative, a few adjustments to how you calculate and apply it will make a significant difference.

Calculate it by acquisition channel, not in aggregate. Blended CLV hides the fact that different acquisition sources produce customers with fundamentally different behaviour. When I’ve run this analysis for clients, the variance between channels is almost always larger than expected. Paid social customers churn faster. Organic search customers often have higher initial intent but lower repeat purchase rates. Referral customers are usually the most valuable by a significant margin. Knowing this changes how you allocate acquisition budget.

Use margin, not revenue. This requires cooperation from finance, which is sometimes harder than the analysis itself, but it’s non-negotiable if you want CLV to be commercially meaningful. A customer generating £10,000 in revenue at 20% gross margin is worth less than a customer generating £6,000 at 50% gross margin. Revenue-based CLV will tell you the opposite.

Cap your projection horizon at a defensible timeframe. For most businesses, projecting customer value beyond three years introduces more noise than signal. The competitive environment changes, your product changes, customer needs change. A three-year CLV with honest margin assumptions is more useful than a five-year CLV built on optimistic churn projections.

Include the cost of retention activity. If you’re spending on loyalty programmes, re-engagement campaigns, or customer success headcount to maintain a customer relationship, those costs should reduce the CLV calculation. If they don’t, you’re overstating the value of customers who require significant investment to retain. Forrester’s intelligent growth model makes a related point about the difference between revenue growth and value-creating growth, and it’s a distinction that CLV models frequently blur.

Segment by behaviour, not just by value. The most useful CLV segmentation I’ve seen doesn’t rank customers from highest to lowest value. It identifies distinct behavioural archetypes: customers who buy frequently at low ticket, customers who buy rarely at high ticket, customers who refer, customers who churn at first renewal, customers who expand. Each archetype requires a different commercial response. A single CLV number per customer obscures these differences.

The Retention Investment Question

There’s a version of CLV thinking that leads to significant overinvestment in retention. The logic is that retaining a customer is cheaper than acquiring a new one, which is generally true, and therefore retention investment has a high return, which is not always true.

The question that rarely gets asked is: what would have happened without the retention activity? Some proportion of customers who receive a retention email, a loyalty reward, or a win-back offer would have stayed anyway. The incremental value of the retention programme is only the customers it retains who would otherwise have left. Everything else is cost with no return.

This is the same measurement problem that exists in performance marketing, where much of what gets attributed to a paid channel was going to happen regardless of the ad. I spent a long time earlier in my career treating lower-funnel performance metrics as proof of marketing effectiveness. It took running proper incrementality tests to understand how much of that activity was capturing intent that already existed rather than creating new demand. Retention programmes have exactly the same problem, and they’re subject to even less scrutiny because the instinct to “keep customers happy” feels obviously correct.

The businesses that have the most honest view of retention ROI tend to run holdout tests: a control group that receives no retention communication, measured against a treatment group that receives the programme. The gap between the two groups is the actual value of the programme. Most businesses don’t do this because the results are uncomfortable. If your retention programme is only marginally better than doing nothing, that’s a difficult conversation to have with the team that built it.

Frameworks for rigorous growth measurement, including how to structure holdout testing and attribution for retention programmes, are covered across the growth tools and analytics resources that practitioners actually use in practice.

CLV and New Customer Acquisition: The Balance Most Businesses Get Wrong

There’s a tendency, particularly in mature businesses, to over-rotate toward retention at the expense of acquisition. CLV thinking accelerates this tendency because it makes existing customer relationships look more valuable than they often are, while making acquisition look expensive by comparison.

The problem is that a business that stops acquiring new customers at scale is a business that’s slowly shrinking its addressable base. Churn is inevitable. Even a business with excellent retention will lose some percentage of its customer base every year. If you’re not replacing those customers and adding net new ones, you’re managing decline, not growth.

More importantly, new customer acquisition is the mechanism by which you reach people who have never heard of you. Retention, by definition, only works on people who already know you exist. Growth requires reaching new audiences, not just serving existing ones better. The businesses that confuse these two things tend to build very efficient retention operations and then wonder why their total customer count isn’t growing. BCG’s research on scaling growth operations points to the same tension: efficiency gains in serving existing customers don’t substitute for the harder work of market expansion.

The right balance between acquisition and retention investment depends on where a business is in its growth cycle, what its churn rate looks like, and what its market penetration is relative to its total addressable market. There is no universal ratio. But the businesses I’ve seen get into trouble are almost always the ones that have let acquisition atrophy because retention metrics looked healthy. CLV often provides the intellectual justification for that mistake.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions