LTV/CAC: Reconstruct and Audit Subscription Unit Economics

Learn how to define customer cohorts, reconstruct acquisition cost and lifetime contribution value, test retention assumptions, calculate payback, and audit LTV/CAC disclosures.

For educational purposes only; not investment advice. Investing may result in loss.

Direct answer

LTV/CAC compares an estimated customer’s lifetime economic contribution with the cost of acquiring that customer. It can help evaluate subscription, software, marketplace, financial-service, and consumer businesses, but only when numerator and denominator refer to matched customers, acquisition periods, cost boundaries, margin definitions, and time horizons.

LTV/CAC is usually a company-defined operating metric, not a standardized accounting ratio. “Customer,” “new,” “acquired,” “churn,” “margin,” and “sales and marketing cost” can differ across issuers and over time. A reported ratio can rise because retention and contribution economics improved, because acquisition spending became more efficient, or merely because management changed definitions, excluded costs, mixed cohorts, or projected an optimistic terminal lifetime.

Read the ratio with cohort retention, gross and contribution margin, CAC payback, net revenue retention, bookings or revenue recognition, cash flow, capitalized contract costs, dilution, and growth capacity. A high modeled lifetime value does not fund today’s acquisition bill, and a short payback estimate does not prove that later cohorts will retain or expand.

How it works

Reconstruct and audit LTV/CAC in this order:

Define the unit and cohort. Specify account, logo, subscriber, seat, order, household, merchant, or user; new versus reactivated or acquired customers; acquisition channel, geography, product, contract, and start date; and whether parent-child accounts, free users, trials, and migrations are deduplicated. Freeze each cohort using information available at the measurement date.
Build fully loaded CAC. A cohort form is CAC_cohort = attributable acquisition spend / acquired customers. Reconcile demand generation, media, sales payroll, commissions, partners, events, software, allocated overhead, trials, implementation, and onboarding to reported expenses and cash payments. State whether the measure is gross, paid-channel, blended, new-logo, incremental, or marginal CAC and how brand and shared costs are allocated.
Measure retention on matched bases. Separate logo churn, gross revenue retention, net revenue retention, contraction, expansion, pauses, reactivation, and involuntary churn. Use exposure-adjusted denominators and comparable intervals. A monthly churn rate cannot be inserted into an annual formula, and annual churn = 12 × monthly churn is only an approximation that ignores compounding.
Estimate customer contribution. Start with recognized or cohort-attributed revenue, then deduct the costs included in the stated margin: hosting, support, payment processing, fulfillment, servicing, revenue share, fraud, credits, and other variable or incremental costs. Gross margin and contribution margin are not interchangeable. Separate historical realized contribution from forecasts of price, usage, cross-sell, and cost-to-serve.
Model lifetime value explicitly. A discounted cohort model is LTV = Σ[Survival_t × Contribution_t / (1 + d)^t] - later customer-specific costs. Under constant end-period contribution m, retention r, and periodic discount rate d, the infinite-horizon simplification is LTV = m / (1 + d - r) when r < 1 + d. With zero discount and churn c = 1 - r, it reduces to LTV = m / c. Do not use the shortcut for nonconstant hazards, contractual terms, negative contribution, finite lives, or expansion that makes the series diverge.
Calculate ratio and payback without mixing units. Use LTV/CAC = LTV / CAC. A no-churn static shortcut is payback periods = CAC / periodic contribution per acquired customer; a cohort payback instead finds the first time cumulative realized or expected contribution, after the stated costs and discounting policy, equals CAC. Disclose whether payback is gross-margin, contribution-margin, cash, or accounting based.
Reconcile, segment, and stress. Tie inputs to filings and cohort tables, preserve definition changes, and compare acquisition cohorts by channel, product, geography, contract size, and maturity. Stress retention curves, margin, discount rate, expansion, CAC allocation, payback, capacity, and terminal horizon; compare forecasts with later realized outcomes rather than relying on one point estimate or a universal threshold.

Revenue recognition and contract-cost accounting do not define economic LTV or CAC. A commission may be capitalized and amortized for accounting while cash was paid near acquisition; billed cash, bookings, remaining performance obligations, and GAAP or IFRS revenue can have different timing. Keep the economic metric, accounting statements, and cash schedule separate and reconcile them explicitly.

Example

Use one acquisition cohort to see how definitions and timing change the conclusion:

CAC: a company attributes $12.0000 million of fully loaded acquisition cost to 3,000 genuinely new customers, so CAC = $12,000,000 / 3,000 = $4,000.00. If it excludes $600,000 of qualifying partner and sales costs, reported CAC falls to $3,800.00; the economics did not improve merely because the boundary changed.
Contribution and simple LTV: each active customer produces $200.00 of monthly revenue at 75.0000% contribution margin, or m = $150.00. With constant monthly churn c = 2.0000% and no discounting, LTV = $150.00 / 2.0000% = $7,500.00, so LTV/CAC = $7,500.00 / $4,000.00 = 1.8750x.
Discounting and payback: with monthly retention r = 98.0000% and discount rate d = 0.7500%, the end-month perpetuity gives LTV = $150.00 / (1.0075 - 0.9800) = $5,454.55 and LTV/CAC = 1.3636x. The static no-churn shortcut is $4,000.00 / $150.00 = 26.6667 months; actual cohort payback is later because customers churn and may never be reached if cumulative contribution is insufficient.
Observed first year: expected survivors after 12 months under the constant hazard are 3,000 × 0.98^12 = 2,354.15, or 78.4717% of the starting cohort. Undiscounted expected contribution during months 1 through 12 is 3,000 × $150.00 × [1 - 0.98^12] / 0.02 = $4,843,873.72, only 40.3656% of the initial CAC. This short observation window does not validate the infinite-horizon retention assumption.

Risks

Define customer, account, subscriber, seat, order, household, merchant, and user consistently.
Separate new, reactivated, migrated, acquired, free, trial, and existing customers.
Match acquisition spend and acquired-customer counts to the same cohort and period.
Reconcile media, payroll, commissions, partners, software, overhead, onboarding, and shared costs.
Distinguish fully loaded, blended, paid-channel, new-logo, incremental, and marginal CAC.
Avoid attributing organic customers to paid spend or crediting one channel for another channel’s conversion.
Separate logo churn, gross revenue retention, net revenue retention, contraction, expansion, and reactivation.
Match monthly, quarterly, and annual rates; do not multiply churn mechanically across periods.
Use mature cohort curves and exposure-adjusted denominators rather than averages distorted by rapid growth.
Distinguish gross margin from contribution margin and disclose every cost excluded from contribution.
Separate historical realized revenue and margin from forecast price, usage, cross-sell, and cost improvements.
Discount future contribution using a stated periodic rate matched to cash-flow timing and risk.
Do not extrapolate a constant churn hazard when retention varies sharply with customer age.
Cap lifetime and terminal assumptions when contracts, products, competition, or technology imply finite economics.
Reconcile bookings, billings, deferred revenue, recognized revenue, contract assets, and cash collection.
Separate accounting capitalization and amortization of commissions from acquisition-period cash economics.
Report payback definition, cohort start, observation cutoff, censoring, refunds, bad debt, and taxes.
Test channel, geography, product, customer-size, contract, and acquisition-vintage mix shifts.
Compare forecast LTV with later realized cohort contribution and disclose definition or methodology changes.
Evaluate growth capacity, working capital, liquidity, dilution, concentration, regulation, and competitive response alongside the ratio.

Common misconceptions

“An LTV/CAC ratio above 3x is always good.” A threshold cannot resolve different margins, horizons, discount rates, capital needs, risks, cohort maturity, or cost boundaries.
“One divided by churn is the observed customer lifetime.” It is a special constant-hazard expectation, not a measured contract duration or a valid shortcut for every retention curve.
“Net revenue retention above 100% creates infinite value.” Expansion can offset churn for a period, but competition, saturation, margin, discounting, customer concentration, and finite market size constrain the forecast.
“CAC is advertising spend divided by sign-ups.” A defensible measure identifies genuinely acquired customers and states which sales, partner, labor, systems, onboarding, shared, and brand costs it includes.
“LTV is a reported asset and high LTV/CAC guarantees cash generation.” Modeled customer value is not a recognized balance-sheet asset, and long payback, cash timing, working capital, taxes, capex, and dilution can still destroy investor value.

Sources

U.S. Securities and Exchange Commission: Non-GAAP Financial Measures Compliance and Disclosure Interpretations.
U.S. Securities and Exchange Commission: MD&A guidance for key performance indicators and metrics.
U.S. Securities and Exchange Commission: How to Read a 10-K.
Financial Accounting Standards Board: Revenue from Contracts with Customers, Topic 606.
IFRS Foundation: IFRS 15 Revenue from Contracts with Customers.
SSRN: Customer Acquisition Cost, Retention and Customer Lifetime Value.