Unit Economics: CAC, LTV and Payback, With the Math

Most startups that claim a 3:1 LTV to CAC ratio are quoting a number that assumes eight years of customer revenue and ignores gross margin. Run the same inputs over a three-year horizon with margin applied and the ratio often falls below 2:1. The arithmetic below shows exactly where the leakage happens, and which of the three numbers actually predicts whether you can afford to grow.

CAC: the numerator is where founders cheat

The formula is trivial. In SaaS Metrics 2.0, David Skok defines customer acquisition cost as sales and marketing expense divided by the number of new customers acquired. The disputes are all about what goes into that expense line.

Everything spent to turn a stranger into a paying customer belongs in the numerator. That means salaries, benefits and commissions for account executives, SDRs and the marketing team; paid media; content and SEO production; events and sponsorships; agency retainers; the CRM, enrichment and sales-engagement tools those teams use; partner and channel commissions; and the cost of any implementation or onboarding work you do not bill for.

Three exclusions matter. Product and engineering costs stay out, even for a self-serve funnel, because you would build the product regardless. Customer success headcount pointed at renewal and expansion belongs in a separate expansion CAC, not new-customer CAC. And support costs sit in cost of revenue, where they reduce gross margin instead.

Two timing errors worth fixing

1. Blended versus new-customer CAC. Blended CAC divides all sales and marketing spend by all new customers including those who arrived organically. It flatters you. Benchmarkit’s 2025 B2B SaaS Performance Metrics report tracks both, and reports a median new-customer CAC ratio of $2.00 of sales and marketing expense per $1.00 of new-customer ARR in 2024, up 14% year over year, with the fourth quartile spending $2.82. Expansion ARR came far cheaper, at a median $1.00.
2. Lag. If your sales cycle runs 90 days, the spend that produced this quarter’s customers happened last quarter. Match spend in period t-1 to customers closed in period t, or the number swings wildly whenever you change marketing budget.

LTV: churn and gross margin, or it means nothing

Skok’s definition is LTV = (ARPA x gross margin %) / monthly churn rate. Both adjustments are load-bearing. Gross margin turns revenue into cash you can actually spend on the next customer. Churn sets the horizon.

Two practical notes. First, use gross revenue churn, not net, for the base LTV, then treat expansion as a separate upside; if you plug a net retention figure above 100% into the denominator you get a negative or infinite lifetime, which is not a number you can budget against. Second, cap the horizon. A monthly churn rate of 1% implies an average customer life of roughly 94 months. Almost no early-stage company has evidence for what happens in year six.

Benchmarkit put 2024 median gross revenue retention at 88%, down from 90% in prior years, with median net revenue retention of 101% and median subscription gross margin of 81%. Retention varies enormously by price point. ChartMogul’s 2025 retention report, drawn from roughly 3,500 software companies, found median gross revenue retention of 70% for products above $250 per month, 45% in the $50 to $249 band, and 23% below $50 per month. A cheap product does not just earn less per customer; it keeps them for a fraction of the time.

For a sense of scale on ARPA, HubSpot reported average subscription revenue per customer of $11,683 in the fourth quarter of 2025 across 288,706 customers and $3.13bn of full-year revenue. That is roughly $975 per month, which is a reasonable mid-market anchor for the example that follows.

The worked example

Take a hypothetical B2B SaaS company. It closed 120 new customers last quarter and spent $1.8m on fully loaded sales and marketing, giving a CAC of $15,000. Average revenue per account is $1,000 per month ($12,000 ACV). Subscription gross margin is 78%. Annual gross revenue retention is 88%, which converts to a monthly gross revenue churn rate of 1.06%.

Same company, same inputs, and the headline ratio moves from 6.7x to 1.6x depending on which conventions you use. The 6.7x version is the one that ends up in board decks. The 1.6x version is the one that tells you the business cannot self-fund growth for at least three years, because every dollar of gross profit a customer generates in that window is nearly all consumed by the cost of winning the next one.

Why the 3:1 rule is weak

The rule has a specific origin. Skok wrote that “our guideline for a successful SaaS business is that this number should be higher than 3,” alongside a separate guideline that months to recover CAC should be under 12, a threshold he dates to 2011 and has since described as more flexible. It was never presented as a law.

“The numbers will only really be meaningful and reliable when you have found a repeatable and scalable growth process.” David Skok, quoted in Why early-stage startups should wait to calculate LTV:CAC (2017).

Four specific weaknesses:

• It is horizon-agnostic. A 3:1 ratio built on a 96-month implied life and a 3:1 ratio built on 24 months describe completely different businesses. The ratio itself does not tell you which you have.
• It hides payback. Two companies can both hit 3:1 with payback periods of 8 months and 30 months. Only one of them can grow without continuous outside capital.
• Early CAC is not real CAC. The same 2017 piece by Jared Sleeper and Skok notes that “as a company needs to scale lead flow, it is forced to add more expensive channels.” Founder-closed deals and warm intros produce a CAC that does not survive contact with a real sales team.
• Retention differs by segment far more than the ratio does. Kyle Poyar’s June 2026 analysis of 1,043 SaaS and AI companies that reached $10k MRR found annualized gross revenue retention of 24.6% for companies selling below $30 per month, 31.6% at $30 to $299, and 70.5% at $300 to $2,999. A blended LTV across those segments is meaningless.

CAC payback is the number to run the company on

Payback in months is CAC / (ARPA x gross margin %). It has three properties LTV:CAC lacks: it needs no assumption about year five, it maps directly onto how long your cash is tied up, and it is comparable across segments.

Benchmarkit reported that median CAC payback rose 12.5% between 2022 and 2024, and that companies with ACVs above $250,000 showed materially shorter payback than those in the $10,000 to $100,000 range. If your payback is 19 months on gross profit, as in the example above, then every customer you add consumes 19 months of that customer’s contribution before funding anything else. Grow new-customer count 50% and the cash hole grows with it.

What to put on the metrics page next quarter

Report four numbers, not one. New-customer CAC with the numerator itemised so anyone can audit it. CAC payback in months on a gross-profit basis. LTV:CAC computed on a stated, capped horizon, with the cap written next to it. And gross revenue retention by price band, because that is what determines whether the horizon assumption holds at all.

Then do one sanity check: multiply your monthly gross-profit ARPA by 24 and compare it to CAC. If that number is below 1, the business is not yet able to pay for its own growth, whatever the ratio on slide 12 says.

Sources

• For Entrepreneurs (David Skok) — SaaS Metrics 2.0: Detailed Definitions
• For Entrepreneurs — Why early-stage startups should wait to calculate LTV:CAC
• Benchmarkit — 2025 B2B SaaS Performance Metrics Benchmarks
• ChartMogul — The SaaS Retention Report: The AI Churn Wave (2025)
• Growth Unhinged (Kyle Poyar) — Why SaaS deal size is your destiny
• HubSpot — Q4 and Full Year 2025 Results