Analytics
CAC payback calculator for marketplaces and two-sided platforms
By Charles Summers · Updated · Free, no signup
Short answer
A marketplace earns its take rate, not its gross transaction volume, so payback is calculated on the commission per active user per month and on the margin left after payment processing, which is charged on the full transaction and therefore eats a much larger share of a take rate than of a subscription price. The second correction is structural: you acquire two sides, the same commission dollar cannot repay both of them, and a platform can show perfectly healthy payback on demand while supply acquisition never repays at all, which is the position where growth quietly destroys value.
Use the cac payback period calculator
Why marketplaces need a different approach
The most expensive mistake in marketplace unit economics is a units error. A platform processing 180 dollars of transactions per buyer per month at a fifteen percent commission has 27 dollars of revenue, not 180, and running payback on the larger figure produces an answer that is wrong by the reciprocal of the take rate, which is to say wrong by a factor of six or seven rather than by a few percent. Everything downstream of that substitution is fiction.
The margin question is subtler and catches operators who got the first part right. Payment processing is levied on the gross transaction, not on the slice you keep, so on a 60 dollar order at fifteen percent you collect nine dollars and hand roughly two of it straight to the processor before anyone has answered a support ticket, refunded a dispute or paid for the guarantee that makes strangers willing to transact. Marketplace gross margin on net revenue is real, but it is nowhere near software margin, and assuming otherwise shortens every payback figure you produce.
It runs entirely in your browser. Nothing you type is sent to a server, no account is required, and there is no usage limit, because there is no cost per run to control.
Payment processing is charged on the transaction, and it lands on the take rate
Work an order through. A 60 dollar booking at a fifteen percent commission gives the platform nine dollars of revenue. Card processing at the common rate of around 2.9 percent plus a fixed thirty cents is levied on the whole 60, so it costs 2.04, which is nearly 23 percent of the nine dollars you actually earned. Nothing about that is unusual or badly negotiated; it is arithmetic that follows from taking a percentage of someone else's revenue while paying fees on all of it. Halve the take rate and the same fee consumes nearly half your revenue.
The other costs that sit in a marketplace cost of goods follow the same pattern of being sized by transaction rather than by revenue. Refunds and chargebacks, dispute handling, identity verification, insurance or a purchase guarantee, and the support burden that arrives whenever two strangers disagree about what was delivered. Trust is a cost of goods in this model, not a marketing expense, and platforms that categorise it as overhead report a margin their payback cannot deliver on.
Which is why the margin figure to enter here is usually well below the number a software business would use, and why the sensitivity runs the other way. In subscription software, price is the fastest lever on payback. In a marketplace the equivalent lever is the take rate, and it is bounded by what the supply side will tolerate before listing elsewhere or transacting off-platform. That ceiling is a competitive fact about your category rather than a pricing decision you get to make, so improving marketplace payback usually means raising transaction frequency or basket size instead.
Two sides, one commission, and the double-count that hides a broken platform
Both sides have to be acquired, at least at the start, and each has its own cost and its own retention behaviour. The temptation is to run this calculator twice, once for buyers and once for sellers, crediting the full take rate to each. That produces two comfortable paybacks and one broken platform, because the nine dollars from the order above is the same nine dollars in both models. A transaction only happens when an acquired buyer meets an acquired seller, so the honest test adds the acquisition cost of both sides, amortised over the transactions each of them generates, and compares the total to the margin on a single commission.
Run separately for diagnosis, added for judgement. The separate figures tell you which side is expensive and which is retaining; the combined figure tells you whether the platform is viable. Most marketplaces find the two sides behave nothing alike. Supply is often expensive to win once and then produces transactions for years, giving a long payback attached to a long lifetime. Demand is cheaper per user and far more fickle, giving a short payback attached to a short lifetime. Neither pattern is a problem on its own. The combination that kills platforms is a demand side that repays quickly and a supply side that never does, because the growth team can hit every target while each additional cohort widens the hole.
The allocation question underneath this is genuinely hard, since much of what a marketplace spends serves both sides at once. Category-level brand spend brings sellers and buyers. Search visibility on a listing page acquires demand while advertising the platform to supply. There is no clean split; there is only a rule stated in writing, applied consistently, and reported with the rule attached, so that a change in the number means something changed in the business rather than in the accounting.
Liquidity decides whether an acquired user produces anything at all
Every figure in this calculation assumes an acquired user transacts at the average rate. In a marketplace that assumption holds only where the market is liquid. Acquire buyers into a category, city or route with insufficient supply and they search, find nothing worth booking, and leave; they arrive in your cohort as acquisition cost with no transaction attached, which pushes the average revenue down and the churn rate up simultaneously, and the payback figure for the whole platform degrades for reasons that have nothing to do with marketing efficiency.
This is the sense in which marketplace payback is not really a marketing metric. It is a report on match rate. Before spending on the abundant side, the useful diagnostics are the share of searches that return a viable result, the share of listings that transact within a given period, and how both vary by geography and category, because acquisition into an illiquid pocket has a payback of never regardless of what the blended figure says. Spend on the constrained side has a multiplier the formula cannot see, since one additional seller can unlock transactions for buyers you already own and already paid for.
Leakage is the last mechanism worth separating out, and it is invisible in the churn number. When a buyer and a seller meet on the platform and then continue the relationship off it, both sides look retained on their profile and dead in the transaction file, and the churn rate reports a behaviour change that is really an escape. It concentrates in categories with repeat relationships and high-value bookings, which are also the categories where a match is worth the most. The remedies are different in kind from churn work: making payment, guarantee, scheduling and dispute resolution genuinely easier on-platform than the alternative. Treat leakage as its own line before you accept a churn input, or you will apply retention tactics to an economic decision your users made deliberately.
Numbers worth knowing
| Metric | Typical | What it means |
|---|---|---|
| The figure that belongs in the revenue field | take-rate revenue per active user per month | Not gross transaction volume. Using GMV overstates the answer by the reciprocal of the take rate, which at fifteen percent means the payback figure is out by a factor of about seven. |
| Processing cost against a take rate | about 23% of revenue on a $60 order at 15% | Fees are charged on the transaction, not the commission: 2.9% plus 30 cents on $60 is $2.04 against $9.00 of revenue. Lower take rates make the share worse, not better. |
| Cost of goods for a marketplace | trust, disputes, guarantees, verification | These scale with transactions rather than with revenue, so they belong in the margin figure. Platforms that book them as overhead report a margin their payback cannot support. |
| Payback on the abundant side | uninformative, not merely poor | A user acquired into an illiquid pocket does not transact at the average rate, so the figure is reporting match rate rather than acquisition efficiency. |
Mistakes that quietly cost you results
- Running payback against gross transaction volume
- You never receive that money; you receive the commission on it. Enter take-rate revenue per active user per month, or the resulting payback is short by one divided by your take rate, which is an order of magnitude rather than a rounding error.
- Crediting the same commission to both the buyer and seller model
- One transaction produces one take-rate dollar. Compute each side separately to see which is expensive, then add both acquisition costs against a single commission to see whether the platform actually works.
- Buying more of whichever side is cheaper to acquire
- Acquisition into the abundant side buys searches with no results. The constrained side has a multiplier the formula cannot see, because one new supplier unlocks transactions from demand you have already paid for.
- Reading off-platform leakage as ordinary churn
- Those users did not lose interest, they took the relationship elsewhere after you introduced them. It concentrates in repeat, high-value categories, and it responds to payments, guarantees and scheduling being better on-platform, not to a win-back campaign.
What does the output look like?
This is the exact output the tool produces from the example inputs. It is generated by the same code that runs when you click the button, so what you see here is what you get.
Frequently asked questions
Do I enter gross transaction volume or take-rate revenue in the revenue field?
Take-rate revenue, meaning the commission you actually keep from an active user in a month. A buyer transacting 180 dollars a month on a platform charging fifteen percent contributes 27 dollars, and entering the 180 makes the payback figure roughly seven times too optimistic. If your commission varies by category or by seller tier, use the weighted average commission that your own transaction mix produced last quarter rather than the rate card.
What gross margin should a marketplace use, given software companies quote such high numbers?
Considerably lower, because the costs that scale with transactions are levied on the transaction while your revenue is only a slice of it. Payment processing is the clearest case, but disputes, refunds, identity verification, insurance and purchase guarantees behave the same way. Build the figure from your own commission downward rather than borrowing a software benchmark, and expect the answer to be uncomfortable relative to the margins your investors are used to seeing.
How should acquisition spend be split between the supply and demand sides?
With a written rule applied consistently, because there is no correct answer. Category brand spend, listing-page visibility and most content serve both sides simultaneously. What matters is that the split does not get renegotiated whenever the resulting number is awkward, since a rule that moves makes the time series a record of accounting decisions instead of a record of the business. State the rule wherever the payback figure is published.
Which side's payback should we be optimising?
The constrained one, which is usually not the expensive one. If demand outstrips available supply, every dollar spent on demand acquisition buys empty search results, while a dollar spent on supply unlocks transactions from buyers you already own and already paid for. That inverts as liquidity improves, so the answer changes by category and by geography and needs rechecking rather than setting once.
Can a marketplace have a healthy payback on one side and a fatal one on the other?
Routinely, and it is the failure mode most likely to survive several quarters undetected. Demand acquisition repays quickly, everyone reports on demand acquisition, and supply acquisition sits in a different team's budget with a payback measured against a lifetime nobody has observed. The platform-level test is the sum of both sides' acquisition cost against the margin on the commission they jointly produce, and it is the only version of this calculation that can tell you the platform works.
Related free tools
- CAC Payback Period Calculator Analytics
- CAC Payback Period Calculator for saas Analytics
- CAC Payback Period Calculator for ecommerce Analytics
- CAC Payback Period Calculator for agencies Analytics
- LTV to CAC Ratio Calculator Analytics
- Net Revenue Retention Calculator Analytics
- Pipeline Coverage Calculator Analytics
- Cold Calling Scriptwriter for recruitment agencies Outbound
- Schema Markup Generator for article pages SEO
Some links on this site are affiliate links, which means Hacking Demand may earn a commission if you buy through them at no extra cost to you. This does not influence which tools are listed. The tools on this page are free and have no affiliate relationship of any kind.