Short answer. An EBITCAC schedule is the reconciliation a CAC lender works from. It walks from the sales and marketing line on your P&L to the acquisition cash actually paid, then allocates that cash to monthly customer cohorts. In CVF's diligence we rebuild its key lines from your general ledger, CRM, and billing exports.

The request usually arrives as one line in a diligence list: substantiate your customer acquisition cost. What that means in practice is a schedule where every number has a source system behind it. In CVF's CAC-financing process, the schedule is a core diligence workpaper after the initial financial screen: we tie it to your P&L, recompute selected cohorts, and sample individual transactions rather than take the summary on faith.

Why the S&M line is not your CAC

The sales and marketing line on a P&L is an accounting bucket, not an economic one. It mixes money spent winning new customers with money spent keeping and growing existing ones, and it misses acquisition costs that accounting rules place somewhere else entirely.

Inside the line sits spend that has nothing to do with new logos: brand and PR work with no pipeline attached, customer success compensation, the marketing that drives renewals and expansion.

Outside the line sit three real acquisition costs. A first-year promotional discount is not an expense at all under revenue-recognition rules: it reduces the transaction price, so it never reaches S&M. Upfront partner and referral fees often sit in cost of revenue. And sales commissions whose benefit runs past a year are required to be capitalized and amortized (a practical expedient lets companies expense them only when the amortization period is a year or less), so the commission expense in any given month largely belongs to deals closed before it.

The EBITCAC framework treats new-customer acquisition as capital expenditure. The schedule is where that idea has to survive contact with your books, and the first thing it must do is stop treating four different totals as one number: booked S&M expense, new-logo acquisition cost, acquisition cash paid, and the lender-eligible base. Accrual accounting separates the first two; payment timing separates the third; credit policy defines the fourth.

The bridge: from booked S&M to the eligible base

The bridge starts at the P&L, subtracts what is not acquisition, adds what is booked elsewhere, converts to cash, and only then applies eligibility. Each subtotal is its own line, because each answers a different question.

StepLineSource
1Booked S&M expense, per P&LGeneral ledger
2minus: retention, expansion and customer success compensationPayroll, employee-level allocation
3minus: brand, PR and events without pipeline evidenceLedger detail, campaign records
4plus: upfront partner and referral fees for new logos booked in cost of revenuePartner contracts, AP
5plus: commissions earned on new deals, attributed per dealCommission records, or payroll register plus comp plan
6= New-logo acquisition cost, accrual basisDerived
7plus or minus: cash-timing adjustments (AP and accrual movements, prepaids, commission payable roll-forward, clawbacks)AP, bank, payroll
8= Acquisition cash paid in the spend monthDerived
9minus: amounts unsupported by exports or ineligible under CVF policyDiligence
10= CVF-eligible cost baseDerived
11Memo: first-year discounts, a non-cash pricing concessionBilling
12Control: unmapped S&M ledger accounts, zero or explainedLedger map

The memo line deserves a sentence of its own. A discount is money you agreed not to receive, not money you paid, so it never enters the cash rows. It is already reflected in the denominator through net MRR, and counting it in CAC as well would charge the same concession twice. At CVF we read a heavy discount memo as pricing pressure for the credit discussion, not as spend.

The control line is what makes this a reconciliation rather than a list. Every S&M ledger account lands somewhere in the bridge: kept, excluded, or explained. One unmapped account is enough to break the tie to the P&L.

Spend month and cohort month are different clocks

The bridge above runs on the spend month. Cohort economics run on the cohort month, which at CVF means subscription activation date. The schedule needs both clocks and an explicit rule connecting them.

Dividing January's spend by January's new customers compares money spent chasing tomorrow's customers with customers won by yesterday's money. In a company with rising spend that overstates CAC; with falling spend it understates it. The exact distortion depends on your conversion curve, which is the point: the naive ratio moves with your budget trend, not with your economics.

Allocation follows the nature of the cost. Commissions and partner fees attach to deals directly, by deal ID. Pooled spend, media and salaries, is spread over the conversion lag observed in your own CRM timestamps. At CVF we accept whatever lag your funnel data supports; what we do not accept is a lag that changes between versions of the document to make a quarter look better. Method changes apply prospectively, with the version noted.

The control here: allocated plus pending plus unallocated equals the cash pool. Money is allowed to wait for its cohort; it is not allowed to disappear.

Where each line comes from

The skeleton is three systems: the general ledger for spend, the CRM for deals and dates, billing for MRR and effective prices. Real companies fill the skeleton with fallbacks, and the schedule states which ones.

No dedicated commission ledger? A payroll register plus the comp plan plus deal IDs does the job. No partner ledger? Contracts and AP invoices do. Ad platform spend reconciles three ways, delivery report to vendor invoice to payment, because a platform dashboard is neither an invoice nor cash.

Payroll needs employee-level allocation with a stated basis: comp plan, commission credits, or documented time. A role label alone is not a basis, since one AE can carry new business and renewals at once. The notes also state one decision founders skip: whether employer taxes and benefits ride along with the salaries they burden.

MRR needs its normalization rules written down: annual prepay, contractual ramps, free months, currencies. The result reconciles back to the billing population, and it carries one honesty label everywhere it appears: MRR at activation is a run-rate figure, not cash collected.

None of this requires an audit. The schedule is a management document, and what an underwriter needs is not a signature but a chain of provenance they can walk. A modest schedule where every figure opens back into an export beats a polished one that exists only in a slide.

What CVF tests when re-performing a schedule

Reclassification comes second. First we tie the schedule to the ledger and check that the CRM and billing populations are complete, with no duplicates and no missing records across the join.

Then the line moves, in rough order of frequency:

  • Renewal and expansion compensation moved out. Caught by employee-level mapping against comp plans, not job titles.
  • Amortization swapped for the commission roll-forward. Earned, capitalized, amortized and cash paid are shown as four fields, with clawbacks in the roll-forward, so no cohort borrows cost from another.
  • Upfront partner fees pulled in; ongoing revenue share left out. A lifetime share of revenue is a margin item for the cohort, not a one-time acquisition cost.
  • One-off launch spend held to its pre-set rule. An allocation backed by pipeline evidence stands. Without evidence, the amount stays in total economics but out of the eligible base, and it stays visible in the reconciliation rather than quietly spread.

The metric-level mistakes, blended CAC, book ARPA, blended margin, are a separate subject covered in our guide to the CAC payback lenders want before funding growth. The schedule comes first: it decides whether there is a number worth reading at all.

What the schedule gets you, and what it does not

A clean schedule makes your acquisition spend reviewable. It does not by itself make the spend financeable, and it does not commit anyone to terms.

At CVF, spend that survives the bridge may qualify for inclusion in our eligible-cost base, subject to full underwriting and facility terms. Retention, concentration, collections and the rest of the credit picture still decide the outcome; the schedule decides whether the conversation has a floor.

Our channel rule at CVF: a channel enters the base once the schedule shows at least one full payback cycle of observed history for it, in billing rather than in projection. Until then, the channel is the company's bet, not the lender's. Whether the resulting numbers are any good, and what a good EBITCAC looks like by stage, is a separate question from having a document an underwriter can stand on.

The one-page summary, and what sits behind it

One row per cohort month, the six most recent vintages on the page, and enough history behind the page to show at least one cohort through full payback.

The summary columns: cohort month; the spend window it draws from; booked S&M; net reclassifications; cash-timing adjustments; eligible cash; unallocated balance; new logos activated; net MRR at activation, labeled as run-rate; the margin basis used; months of maturity; and payback, in both of its forms.

Margin honesty first. Few early-stage companies can compute a cohort's own gross margin, because that needs hosting, support and payment costs at customer level. The schedule allows three levels, actual cohort margin, product or segment proxy, or company-wide proxy from our gross margin guide's definitions, and states which one each row uses. A proxy is fine; a proxy presented as a measurement is not.

Payback comes in two forms, and they answer different questions. The run-rate proxy divides eligible cohort cost by net starting MRR times the margin basis: available immediately, and it moves if churn arrives later. Observed payback is the month when the cohort's cumulative collected gross profit, net of refunds, crosses its cost: slower to arrive, but it is the number repayment actually follows. The formula mechanics live in our CAC payback guide; the schedule's job is to keep the two forms from masquerading as each other.

Behind the one page sit the working tabs: the ledger map, the cash bridge, payroll and commission detail, the CRM-to-billing join, the cohort allocation table, and an exceptions list. Six vintages fit on the page; the file behind it reaches back through a full payback cycle, with maturity dates on every row so a three-month-old cohort is never read as a finished one.

When the data is thin: the minimum reviewable version

Shrink the claim, not the honesty. The minimum reviewable version separates what is supported from what is not, instead of blending them into one confident number.

It shows three totals. A fully loaded upper-bound CAC, with everything plausibly acquisition included. A supported base, holding only what traces to exports. And an unresolved remainder, listed item by item. Disputed amounts sit outside the supported base until evidenced, because an inflated base is not conservative from the lender's side of the table: it overstates exactly the number the advance would be sized on.

Thin and traceable reads as an early-stage company. Precise and unsourced reads as a red flag. And thin data has consequences either way: at CVF it narrows the eligible base or adds a haircut, and it can be the reason an application waits for another quarter of history.

Frequently asked questions

What is an EBITCAC schedule in one sentence?

A reconciliation from the S&M line on your P&L to acquisition cash paid, allocated to monthly customer cohorts, with every figure traceable to the general ledger, CRM, or billing.

Is the EBITCAC schedule an audited document?

No. It is management-prepared. An audit helps, since the P&L the schedule ties to is then itself verified, but a lender relies on re-performing the key lines from raw exports, not on a signature.

Do first-year discounts belong in CAC?

As a memo line, not in the cash rows. A discount is revenue you agreed not to receive: a pricing concession, not spend. It already flows through the denominator via net MRR, so adding it to CAC as well would count the same concession twice.

Should commissions be counted as cash or as amortization?

Neither alone: the schedule shows earned, capitalized, amortized, and cash paid as four fields, with clawbacks in the roll-forward. Revenue-recognition rules generally require capitalizing commissions whose benefit runs past a year (KPMG's revenue recognition handbook covers the mechanics); the schedule undoes that smoothing on purpose, and says so in the notes.

How many months should the schedule cover?

Six most recent monthly vintages on the summary page, which is CVF's practice, with supporting history deep enough to show at least one cohort through full payback. A young cohort is not a bad cohort, but its payback column reads immature, not a number.

This article describes CVF's underwriting methodology and the document formats we work from. Booking practices vary by company and jurisdiction; confirm the treatment of commissions, discounts, and partner payouts with your accountants. It is not investment, legal, tax or accounting advice, not an offer to lend or invest, and not a promise of eligibility or approval. CVF makes no assurance of eligibility, approval or funding.