Blog · Finance metrics and formulas
Calculate an ecommerce LTV to CAC ratio using observed customer cohorts and contribution after returns and variable costs. A worked example separates first-order value, repeat purchases and acquisition spending without using a subscription churn formula.
The LTV CAC ratio for ecommerce compares the contribution earned from acquired buyers with what it cost to acquire them. For a business without subscriptions, start with observed first and repeat orders from a fixed customer cohort. State the observation window so value earned so far is not presented as a complete lifetime forecast.
Observed contribution LTV = (Cohort net revenue - Net COGS - Other included variable service costs) / Original acquired buyers
CAC = Cohort acquisition spending / Original acquired buyers
Observed LTV to CAC ratio = Observed contribution LTV / CAC
Because both per-buyer figures use the same denominator, the ratio also equals total cohort contribution divided by cohort acquisition spending. That identity is a useful check against customer counts that change between worksheets.
Contribution here is before acquisition spending. Subtracting acquisition costs in LTV and dividing by CAC would charge the same spending twice in the decision comparison. Document the variable costs you include so finance can reconstruct the measure.
The general LTV to CAC guide explains the broader ratio. This version addresses the non-subscription retailer: irregular repeat orders, delayed refunds and customer relationships without a clear cancellation date.
Research on repeat buying in noncontractual settings treats customer departure as unobserved. That distinction matters in practice. An inactive retail buyer is not automatically a canceled subscriber, so dividing an average margin by a subscription churn rate is not a reliable shortcut for this cohort calculation.
Take the customers whose first completed purchases fall in one acquisition period. Verify earlier purchase history before labeling them new. Keep a stable customer ID, first-order date, acquisition channel and original cohort membership.
Then attach every qualifying order through the same age after each buyer's first purchase. Your order export needs order date, sales after discounts, refund date, refund amount, related net COGS and the variable costs you intend to deduct. Use order IDs to connect later refunds to the original purchase.
Net COGS means that any inventory cost recovered through a return is already reflected. Keep shipping subsidies, payment processing and other order service costs outside that line if your ledger does; include them separately in the contribution bridge. Map each expense once.
Use a separate expense export for acquisition costs, with shared costs allocated on a documented basis. The ecommerce CAC calculation covers that buyer and spending reconciliation in detail.
A US personal-care retailer acquires 1,000 new buyers in a single cohort. Each buyer has complete records through day 180 after the first order. Acquisition spending is assigned to this cohort as follows:
| Acquisition cost | Amount (USD) |
|---|---|
| Media | 35,000 |
| Creative | 5,000 |
| Agency | 5,000 |
| Allocated acquisition staff | 5,000 |
| Total | 50,000 |
CAC is 50,000 / 1,000 = $50.00 per original acquired buyer. Retention spending is not included in that acquisition total; attributable variable retention costs, if any, belong in the service-cost definition used below.
The cohort generates these sales and costs. Sales exclude sales tax and are after discounts. Other variable costs in this example include payment processing, outbound fulfillment, return handling and attributable retention expense.
| Orders | Order count | Sales (USD) | Refunds (USD) | Net revenue (USD) | Net COGS (USD) | Other variable costs (USD) | Contribution (USD) |
|---|---|---|---|---|---|---|---|
| First orders | 1,000 | 100,000 | 10,000 | 90,000 | 45,000 | 15,000 | 30,000 |
| Repeat orders, days 1-90 | 400 | 44,000 | 4,000 | 40,000 | 18,000 | 6,000 | 16,000 |
| Repeat orders, days 91-180 | 300 | 36,000 | 3,000 | 33,000 | 13,500 | 4,500 | 15,000 |
| Total | 1,700 | 180,000 | 17,000 | 163,000 | 76,500 | 25,500 | 61,000 |
The first-order contribution is 100,000 - 10,000 - 45,000 - 15,000 = $30,000. The repeat blocks add $16,000 and $15,000. These are order counts, not distinct repeat-buyer counts: one buyer might place several repeat orders.
IRS Publication 334 sets out the gross-profit bridge of receipts less returns and allowances, then less COGS. The additional subtraction of the explicitly listed variable service costs turns this example into a management contribution measure. It should not be confused with a financial statement's gross profit line.
The observed contribution LTV is 61,000 / 1,000 = $61.00. Divide by CAC of $50.00 to get 1.22 times, displayed to two decimal places.
The direct total check is 61,000 / 50,000 = 1.22. It agrees because the original acquired-buyer count is unchanged. Dividing contribution by 1,700 orders instead would produce a per-order measure and break the comparison with per-customer CAC.
The contribution reconciliation is equally important:
61,000 = 180,000 - 17,000 - 76,500 - 25,500
Contribution after acquisition spending = 61,000 - 50,000 = 11,000
The remaining $11,000 is before fixed overhead, financing and tax. It is not net profit. Reconcile the sales and refunds to the cohort order register, COGS to the related cost records, variable costs to their expense schedule and acquisition spending to its allocated ledger total.
The first orders earned $30,000 of contribution and repeat orders earned 16,000 + 15,000 = $31,000. Keeping the cohort intact shows exactly how repeat purchases moved acquisition recovery. A separate customer lifetime value calculation explains how to distinguish this observed amount from a projection of future purchases.
Suppose the final repeat-order block has an additional $6,000 of refunds, with no recovered inventory or variable-cost refunds. All acquisition costs, buyers and other inputs stay the same:
| Measure | Original | Revised |
|---|---|---|
| Additional refunds (USD) | 0 | 6,000 |
| Final-block refunds (USD) | 3,000 | 9,000 |
| Final-block net revenue: 36,000 less refunds (USD) | 33,000 | 27,000 |
| Final-block contribution: net revenue less 13,500 COGS and 4,500 variable costs (USD) | 15,000 | 9,000 |
| Total contribution: 30,000 + 16,000 + final block (USD) | 61,000 | 55,000 |
| Original acquired buyers | 1,000 | 1,000 |
| Observed LTV: contribution / buyers (USD) | 61.00 | 55.00 |
| CAC: 50,000 / buyers (USD) | 50.00 | 50.00 |
| LTV / CAC (times) | 1.22 | 1.10 |
| Contribution less 50,000 acquisition spending (USD) | 11,000 | 5,000 |
The revised ratio is 55 / 50 = 1.10 times, with $5,000 left before fixed costs. A change in returned sales can remove a substantial part of that cushion.
This is why a revenue multiple is a different result. Original cohort net revenue divided by acquisition spending is 163,000 / 50,000 = 3.26 times. It leaves out both product and service costs. Calling that contribution LTV to CAC would conceal the economics shown by the 1.22 ratio.
Finance owns cost definitions and reconciliations. Marketing owns acquisition assignments and proposed spending changes. Ecommerce operations owns the explanation for refunds, fulfillment costs and order behavior. Together, they should approve the cohort window before comparing channels.
The result helps identify which cohorts deserve a retention review or a controlled acquisition test. Compare cohorts at the same age and inspect the timing of contribution. A positive amount earned late can still require substantial cash funding before recovery. The unit economics guide puts that timing question beside the customer margin calculation.
Bring a sample of your own customer order history, refund records, variable-cost schedule and acquisition expense file to Covirage. Covirage maps the columns, and its tools compute and check the cohort figures; the external AI model explains the result with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring your files.
See customer profitability for the wider view of what acquired customer groups earn and cost.
Ordinary ecommerce does not have an explicit subscription cancellation event. A customer who has not reordered may still return later. Start with observed contribution from equally mature acquisition cohorts. If you later project future value, state the assumptions separately and check predictions against subsequent purchase and margin results.
Include repeat orders from the original acquired customers within the stated observation window. Keep the original cohort size as the denominator, including buyers who never reorder. Removing nonrepeat buyers would measure value among repeaters rather than the value of acquiring an average customer in that cohort.
It shows observed contribution exceeds acquisition cost on the defined basis. It does not establish that fixed costs are covered, cash arrived quickly enough or additional buyers will have the same economics. Review timing, cohort maturity and incremental acquisition costs before increasing a channel's budget.