Contents
In short. The affordable cost of a lead is worked out in two moves. First, the customer's contribution (revenue after refunds and variable costs, including repeat purchases) gives the affordable cost of acquiring a customer. Then the affordable advertising cost per customer is multiplied by the share of leads that reach a sale. In the illustrative example below a customer contributes $150 over six months, $100 can go to advertising per customer, and the affordable cost per lead is $15. This shows a method, not a norm: use the numbers from your own group of orders.
Work from the customer’s money down to the price of a lead
The piece on CPL, CAC and ROAS calculated those measures for inquiries that had already happened. This one solves the reverse problem: before a launch or a budget increase, find the ceiling at which a group of customers still brings in more than it costs to acquire. CPL is cost per lead; CAC is the full cost of acquiring one new customer.
- Define lead, qualified lead, sale, customer and the observation window.
- Contribution per order: order value minus refund minus variable costs.
- Add repeat purchases over the same window to get customer contribution.
- Decide the share of contribution the business keeps; the rest is the affordable CAC.
- Subtract non-advertising acquisition costs to get affordable ad spend per customer.
- Multiply by the share of leads that reach a sale: that is the affordable cost per lead.
- Check when the cash arrives, then test the result on a real group of orders.
How much a customer really leaves the business
Contribution (contribution margin in managerial accounting) is the amount by which the selling price exceeds variable costs. It covers fixed costs first; the rest is profit. Variable costs rise with the number of orders: product or service, shipping, payment processing, commissions, discounts. A manager’s salary and rent are usually fixed and stay out.
Illustrative example. The first order averages $250 and variable costs are $100, so 60% of order value is left. 10% of orders are refunded in full: the money goes back, the variable costs are already spent. Over six months 2 of every 10 buyers buy again, 0.2 repeat orders per buyer, with the same value, refunds and costs. The figures are invented and describe no real company.
Result: first-order contribution is $125 ($250 minus a $25 refund minus $100 of costs), repeat purchases add $25, and customer contribution over six months is $150.
What share of contribution can go to acquisition
That figure, not the whole affordable CAC, is the one that goes on into cost per lead.
Fixed costs and profit are paid from contribution too, so the owner sets a share to keep. In the example it is 20%, or $30, and the affordable CAC is $150 minus $30, which equals $120. The share is a management decision, not a law: the higher your fixed costs and the riskier the channel, the larger the cushion.
CAC is more than advertising. Shopify’s definition of customer acquisition cost includes advertising, tools, marketing pay, discounts, content creation and sales costs that support acquisition. Suppose the non-advertising part in the example is $20 per customer. Then $100 is left for advertising per customer. That figure, not the whole affordable CAC, is the one that goes on into cost per lead.
When the money comes back: payback and the cash gap
CAC payback is the time it takes customer contribution to cover the cost of acquiring that customer. Subscription companies use monthly revenue times gross margin; for ordinary retail it is easier to use contribution in money by month.
In the example, acquisition costs are paid in month zero, and first-order contribution arrives in two parts: $75 in month zero and $50 in month one. At the end of month zero the customer is $45 behind; at the end of month one, $5 ahead. Payback comes in month one, but with 10 new customers a month the temporary gap is about $450.
Profit in a report and cash in the bank differ: the IAS 7 standard prescribes a separate statement of cash flows.
Why the lead ceiling is a fraction of the spend per customer
The funnel has three stages: a lead, a qualified lead (accepted for follow-up under agreed criteria) and a sale (the first payment). 60% of leads qualify and 25% of those end in a sale, so 15% of leads reach a sale. The affordable cost per lead is $100 x 15% = $15; a qualified lead can cost up to $100 x 25% = $25.
A check on 100 leads: advertising is $1,500 ($15 per lead), other acquisition costs are $300, and total acquisition is $1,800. The 15 customers contribute $2,250, leaving $450, which is 20% of contribution, as intended.
| Item | Value |
|---|---|
| Leads | 100 leads |
| Qualified leads | 60 leads |
| Sales | 15 leads |
Illustrative example: 60% of leads qualify, 25% of qualified leads end in a sale.
Scroll right to see all columns
| Step | Calculation | Result |
|---|---|---|
| Refund per order | $250 x 10% | $25 |
| First-order contribution | $250 - $25 - $100 | $125 |
| Repeat-purchase contribution | $125 x 0.2 | $25 |
| Customer contribution, 6 months | $125 + $25 | $150 |
| Share you keep | $150 x 20% | $30 |
| Affordable CAC | $150 - $30 | $120 |
| Affordable ad spend per customer | $120 - $20 | $100 |
| Share of leads that reach a sale | 60% x 25% | 15% |
| Affordable cost per lead | $100 x 15% | $15 |
| Affordable cost per qualified lead | $100 x 25% | $25 |
What moves the ceiling more: conversion or the share left
Before arguing whether an $18 lead is expensive, ask what conversion it requires.
The table recalculates the case for different shares left after variable costs (only variable costs per order change) and different lead-to-sale conversion. The base cell is 60% and 15%, which gives $15.
The ceiling moves in proportion to conversion: at 10% it falls from $15 to $10, at 20% it rises to $20. The share left matters more than it seems: at 40% and the same 15% conversion the ceiling is $7.80, about half the base, because a lower share eats contribution while non-advertising costs stay put.
A practical conclusion: before arguing whether an $18 lead is expensive, ask what conversion it requires. In the base row that price needs 18% instead of 15%, a fifth higher. If your funnel has never shown that conversion, treat an $18 ceiling as risky.
| Item | 40% left | 50% left | 60% left |
|---|---|---|---|
| 5% | $2.6 | $3.8 | $5 |
| 10% | $5.2 | $7.6 | $10 |
| 15% | $7.8 | $11.4 | $15 |
| 20% | $10.4 | $15.2 | $20 |
| 25% | $13 | $19 | $25 |
Horizontal axis: lead-to-sale conversion. Rows of the table below, recalculated from the example conditions.
Scroll right to see all columns
| Share left after variable costs (affordable ad spend per customer) | Lead-to-sale 5% | Lead-to-sale 10% | Lead-to-sale 15% | Lead-to-sale 20% | Lead-to-sale 25% |
|---|---|---|---|---|---|
| 40% ($52) | $2.60 | $5.20 | $7.80 | $10.40 | $13 |
| 50% ($76) | $3.80 | $7.60 | $11.40 | $15.20 | $19 |
| 60% ($100) | $5 | $10 | $15 | $20 | $25 |
| 70% ($124) | $6.20 | $12.40 | $18.60 | $24.80 | $31 |
Which mistakes inflate the ceiling, and by how much
Effects refer to the base example, where the correct ceiling is $15: the order of magnitude of each mistake, not a typical market error.
- Starting from $300 of revenue instead of contribution more than doubles the ceiling.
- A refund removes revenue and often leaves shipping and payment costs already paid.
- Using the advertising budget instead of full CAC gives a good-looking report and a loss-making group.
- Counting repeat purchases on hope means spending against revenue that may never arrive; leaving them out is safer but rejects traffic that would pay off.
- Ignoring payment timing hides a $45 gap per customer, and one average conversion for every channel hides weak segments.
How to test the ceiling on real orders
Take every lead created in one month, not only those that ended in a sale, or you count only the wins. Let the group mature: observation window plus refund window.
Per Google Ads help, conversions can be recorded up to 90 days after the click, so a recent period looks worse; for complete data, end a report at least 30 days before today, longer with a longer conversion window. Refunds can be reflected by reducing or removing a conversion; website conversions need a transaction ID, within a limited period. Yandex Metrica accepts CRM orders with revenue and cost and allows changes for 111 days.
You do not need the ad account for this; an export from your CRM and your books is enough. Build two sheets (replace the ranges with your own; in some locales the argument separator is a semicolon):
- Sheet "Orders", one row per order: A order number; B customer ID; C lead date; D payment date; E first order of this buyer (1 or 0); F amount; G refund; H variable costs; I contribution =F2-G2-H2; J date the money arrived.
- Sheet "Calc", inputs: B1 leads in the group and B2 qualified leads (from the CRM); B4 advertising spend; B5 other acquisition costs; B11 share of contribution you keep (for example 20%).
- Actuals: B3 buyers =SUM(Orders!E2:E500); B6 group contribution =SUM(Orders!I2:I500); B7 contribution per buyer =B6/B3; B8 actual CAC =(B4+B5)/B3; B9 conversion =B3/B1; B10 cost per lead =B4/B1.
- Ceiling: B12 affordable CAC =B7*(1-B11); B13 affordable ad spend per customer =B12-B5/B3; B14 affordable cost per lead =B13*B9.
- Headroom: B15 =B14-B10. Below zero means the group went over the ceiling.
What to do with the result, and where the model can mislead
First compare B8 with B12: if actual CAC is above affordable CAC, the group brought in less than you meant to keep. Then find the cause: conversion fell (B9), cost per lead rose (B10), or contribution and refunds moved (B7). For a budget decision use several monthly groups, and recalculate the ceiling at the worst conversion in the sensitivity table.
Limits. The model uses averages, while order value, refunds, costs and repeat purchases differ by product. The kept share and the window are the owner’s choices, not derived from data. Ask your accountant whether sales tax, VAT and platform fees sit inside order value and costs. The sources confirm definitions and platform behavior but give no industry norms, so we name no "good" CAC or "normal" payback. The calculation guarantees no future sales.
Key decisions
- start from customer contribution rather than revenue, and include refunds and repeat purchases at once
- split the affordable customer cost into advertising and other acquisition costs, and convert only the advertising part into a cost per lead
- test the ceiling on a real group of orders and keep a margin for the case when conversion comes in below expectation
What to discuss with your team
- Lead, qualified lead, sale, customer and observation window are defined
- Contribution is calculated after refunds and all variable costs, with discounts counted once
- Repeat purchases are included only for the window actually observed
- Affordable CAC is split into advertising and other acquisition costs
- The lead ceiling is recalculated at the worst conversion, not only the average
- The result is tested on a real monthly group of leads and against cash timing
Terms in plain language
- Contribution (contribution margin)
- The amount by which the selling price exceeds variable costs; it covers fixed costs first, and the remainder is profit
- Variable costs
- Costs that rise with the number of orders: product, shipping, payment processing, commissions
- Affordable CAC
- The highest full cost of acquiring a customer at which the business still keeps a chosen share of contribution
- CAC payback
- The time it takes customer contribution to cover the cost of acquiring that customer
Sources
- Shopify: Customer acquisition cost, cost scope and the 3:1 guideline (vendor blog, definitions only)shopify.com · checked 31/10/2026
- David Skok, forEntrepreneurs: SaaS Metrics 2.0, LTV:CAC and CAC recovery guidelines (subscription companies)forentrepreneurs.com · checked 31/10/2026
- OpenStax: Principles of Accounting, Volume 2, contribution marginopenstax.org · checked 31/10/2026
- Wall Street Prep: CAC payback period (vendor training material, definition and formula only)wallstreetprep.com · checked 31/10/2026
- Google Ads: About conversion valuessupport.google.com · checked 31/10/2026
- Google Ads: About conversion adjustments (returns and cancellations)support.google.com · checked 31/10/2026
- Google Ads: Find out how long it takes customers to convertsupport.google.com · checked 31/10/2026
- Yandex Metrica: Importing offline data (CRM orders, conversions, calls)yandex.com · checked 31/10/2026
- IFRS Foundation: IAS 7 Statement of Cash Flowsifrs.org · checked 31/10/2026