Customer Lifetime Value: Formula, Examples, and CAC Payback

Customer lifetime value (CLV) is the total revenue or gross profit a business expects to earn from one customer over the full relationship. A common formula is average purchase value × purchase frequency × customer lifespan; the result helps set acquisition and retention spending, but a scaling decision also needs margin and CAC payback.

The basic customer lifetime value formula#

For a first-pass revenue estimate, use:

CLV = average purchase value × average purchase frequency × average customer lifespan

The inputs can be calculated as total revenue ÷ number of purchases, number of purchases ÷ number of unique customers, and sum of customer lifespans ÷ number of customers. HubSpot’s calculation guide and Ramp’s CLV overview present this basic structure. A profit-based variant substitutes gross profit per customer for revenue, which is the more useful boundary when the question is how much contribution can repay acquisition cost.

Here is an illustrative example, not real company data. A subscription-box business has an average purchase value of $50, an average purchase frequency of four purchases per year, and an average customer lifespan of five years. Its revenue CLV is $50 × 4 × 5 = $1,000 per customer.

The same example can be written as a cohort schedule normalized per acquired customer. This makes the time path visible instead of showing only the final multiplication:

This table is a complete numeric revenue-CLV example, but it is deliberately not a CAC-payback claim. To turn it into a decision model, replace the flat illustrative purchase pattern with observed cohort retention, convert each period’s revenue to gross profit under the approved cost policy, enter like-for-like CAC, and find the first period when cumulative gross contribution reaches CAC. Without those inputs, a revenue CLV of $1,000 does not reveal payback.

CLV, LTV, revenue, and profit are not four interchangeable inputs#

CLV and LTV are usually interchangeable names for the same metric. Where writers distinguish them, CLV means the value of one customer and LTV means the average across the customer base; the distinction is a soft convention, not an enforced industry standard. More important is whether the model uses revenue or profit. The basic formula uses revenue, while a profitability view substitutes gross profit per customer.

Historic and predictive CLV also answer different questions. Historic CLV totals what a customer has spent to date. Predictive CLV uses historical and behavioral evidence to forecast future spending. Higher churn lowers either forward-looking estimate because it shortens average customer lifespan.

The widely repeated “good” CLV:CAC ratio is 3:1—about $3 of lifetime value for every $1 of acquisition cost. Ramp and other industry sources repeat this figure, with commonly cited directional ranges of 3:1–5:1 for SaaS and enterprise, 2:1–4:1 for ecommerce, and 2:1–3:1 for fintech.

A ratio below roughly 1:1 is commonly described as unsustainable, while a ratio above roughly 5:1 can indicate under-investment in growth; neither shortcut replaces the cohort economics.

CLV matters because a customer creates economic value only when the value attributed to the relationship exceeds the cost to acquire and serve it. It tells a team how much it may be able to spend on acquisition and retention, but only after the model states its revenue or profit boundary, retention assumptions, and time horizon.

CLV is therefore a forecast, not a fact discovered at signup. The useful question is not “What is our LTV?” in the abstract. It is “For customers acquired in this segment, under these retention and margin assumptions, how much contribution do we expect, when do we recover acquisition cost, and how wrong could the estimate be?” That wording forces the metric to carry the information an acquisition decision actually needs.

Start with a cash-flow definition#

Peter Fader and Bruce Hardie’s critique of common CLV formulas starts from a general definition: expected CLV is the sum of expected net cash flow in each future period, conditional on the customer still being active, multiplied by the probability the customer is active and an appropriate discount factor.

Written for a segment rather than an individual, the structure is:

Expected CLV = Σ [P(active in period t) × expected net cash flow in period t if active × discount factor for period t]

This is a model frame, not a plug-and-play formula. “Net cash flow” needs an explicit cost boundary. “Active” needs a definition suited to the business model. The time horizon and discount rate need owners. If a finite horizon omits meaningful later cash flows, label the output as horizon value or estimate a defensible residual value instead of quietly calling the truncated result lifetime value.

Keep acquisition cost outside this version of CLV. That lets you compare the economic value expected from a newly acquired customer with the cost of acquiring that customer. If you subtract CAC inside CLV and then compare the result with CAC again, you count acquisition cost twice.

Choose the retention model your customer relationship permits#

A contractual business can observe many endings: a subscription is canceled, an account fails to renew, or a term expires. Its model can estimate the probability of remaining active from renewal cohorts. Do not assume that one churn rate applies forever. Early-tenure customers and long-tenure customers can retain differently, and a blended company rate can hide that pattern.

A noncontractual business has a different problem. A customer who has not purchased recently may have left, or may simply be between purchases. In that setting, repeat purchase rate is not an observed survival rate. Estimate future transaction activity from purchase history with a model suited to noncontractual behavior, then combine the expected transactions with expected contribution per transaction.

This choice comes before arithmetic. Using 1 ÷ churn to infer lifetime assumes a stable contractual retention process. Applying that shortcut to irregular repeat purchases gives a precise-looking answer to the wrong model.

Build retention from acquisition cohorts#

Group customers by acquisition period and by the segment that drives economics: plan, market, channel, geography, or another material distinction. Keep the segment definition stable long enough to observe it. A useful cohort table has one row per acquisition cohort and one column per customer age period, with both customer status and revenue carried forward. That analysis depends on a stable customer data contract, not only a dashboard formula.

For recurring-revenue businesses, keep three views separate:

ChartMogul’s retention definitions calculate these revenue metrics from customers present at the beginning of the period and exclude new customers acquired during it. That exclusion is essential: adding new business would make acquisition look like retention.

GRR = (starting recurring revenue − contraction − churn) ÷ starting recurring revenue

NRR = (starting recurring revenue + expansion − contraction − churn) ÷ starting recurring revenue

Do not substitute NRR for the probability that a customer survives. Expansion can lift revenue retained even while accounts disappear. The CLV schedule needs both: logo survival to estimate whether a relationship remains active, and revenue retention to estimate how the value of surviving relationships changes.

Put gross margin on the same rows#

Revenue lifetime value answers how much a cohort may bill. It does not answer how much value remains after delivering the product or service. For an acquisition decision, translate cohort revenue into gross profit or a more tightly defined contribution measure.

For each customer-period row, begin with recognized or collected revenue according to the model’s stated purpose, then subtract the costs included in the chosen margin policy. Reconcile that policy with finance. Hosting, payment processing, implementation, support, fulfillment, returns, and service labor can behave differently across businesses and may be classified differently in their accounts. The model should name which costs it includes rather than relying on the label “gross margin” to settle the question.

Bessemer Venture Partners’ cloud economics framework defines gross margin as gross profit divided by revenue and notes that cloud cost of goods sold commonly includes delivery-related costs such as hosting, implementation, and services. The relevant lesson is not to borrow another company’s margin. It is to use the costs that actually change the contribution from your segment.

Apply margin at the same granularity as retention. A company-wide margin multiplied by segment revenue can misprice a service-heavy customer group, a payment-intensive product, or a channel that attracts a different plan mix. When granular cost allocation is not reliable, show the uncertainty as a range and identify the shared costs that were not allocated.

Derive payback from the same cash-flow schedule#

CLV and CAC payback answer different questions. CLV estimates total future economic value under a forecast. Payback identifies when cumulative customer contribution recovers acquisition cost. A segment can have attractive long-run CLV and still have a payback period that strains cash.

Define CAC for the same cohort and segment as the lifetime-value estimate. Document which sales and marketing costs are included, how shared costs are allocated, how new customers are counted, and which period’s spending produced the cohort. Then calculate:

CAC payback period = earliest period t when cumulative gross contribution through t ≥ CAC

Use actual timing rather than dividing CAC by a steady-state monthly average when onboarding, seasonal purchases, annual prepayments, refunds, or ramping usage make cash flows uneven. Decide with finance whether the operating view uses undiscounted contribution, discounted contribution, or both, and keep that policy fixed across comparisons.

Payback under a gross-margin convention is not the date the whole company becomes profitable. It does not automatically cover research, administration, financing, or every retention expense. Name it precisely: gross-margin CAC payback, contribution payback, or another term that matches the costs in the numerator.

Assemble one auditable model#

The model can live in a warehouse, notebook, or spreadsheet. Its first version needs traceable inputs more than sophisticated software.

Produce four outputs for every material segment: an observed retention curve, expected contribution by customer age, cumulative contribution against CAC, and expected CLV with a downside range. Backtest the forecast on mature historical cohorts: hide later periods, forecast them from the earlier data, and compare the forecast with what subsequently occurred. Record the error and revise the model when the customer mix or pricing changes.

Require a scaling gate, not a flattering ratio#

An LTV-to-CAC ratio compresses timing, uncertainty, and cohort composition into one number. Keep the ratio as a summary, then make the acquisition decision from the underlying evidence.

Before increasing spend, require all of these conditions:

• Retention evidence comes from cohorts old enough to reveal the relevant renewal or repurchase behavior. Immature periods are marked as forecasts, not treated as observed lifetime.
• Margin is based on the segment’s delivery economics and reconciles to the approved finance definition.
• Payback fits the company’s cash constraints under a downside case, not only the average forecast.
• CLV and CAC use the same customer, segment, channel, currency, and cost boundaries.
• The model still supports acquisition when retention, expansion, margin, or CAC moves adversely within a documented plausible range.
• Recent cohorts do not show deterioration that a blended historical average conceals.

If one condition fails, the next task is specific. Repair retention before paying to refill a leaking cohort. Repair pricing or delivery cost when revenue survives but contribution does not. Repair channel economics when the same product retains well but CAC or customer mix changes. Repair cash planning when long-run value is credible but payback is too slow.