SaaS Unit Economics: CAC, LTV, and Payback With Worked Examples

Unit economics answers a narrow question: does a new customer, on average, return more gross profit than it cost to win them, and how long does that take? Three numbers carry most of the answer. Customer acquisition cost (CAC) is what you spend to win one customer. Lifetime value (LTV) is the gross profit that customer produces before leaving. CAC payback is the number of months until the first catches up with the second; public companies such as Similarweb define it as the time needed to recover CAC from the incremental gross profit of newly acquired customers.

The formulas are simple. Most errors come from the inputs, and from one formula in particular, simple CAC payback, which ignores churn and can make a business that never breaks even look healthy. This guide shows that with numbers.

Getting the inputs right

CAC: count everything

CAC = total sales and marketing cost in a period ÷ new customers won in that period.

The denominator is easy. The numerator is where companies flatter themselves. It should include:

• Salaries, commissions, bonuses, and benefits for sales, marketing, and sales development
• Ad spend, agencies, events, content production, and sponsorships
• Sales and marketing software (CRM, sequencing, analytics, intent data)
• Onboarding or implementation work that is needed to close the deal

Two refinements matter once the business has some history. First, if your average sales cycle is four months, compare this quarter's new customers with last quarter's spend, not this quarter's. Second, report CAC by channel and segment as well as blended. A blended $1,500 CAC can hide a paid-social channel at $4,000 and an organic channel at $600, and the average tells you nothing about where the next dollar should go.

Gross margin: use the real one

LTV and payback should use gross profit, not revenue. Hosting, third-party APIs, payment processing, and the support and customer success staff who keep accounts running all belong in cost of revenue. Snowflake's 10-K, for example, counts cloud infrastructure (including GPU and AI inference costs) and customer support personnel in its cost of product revenue. A company that quotes an 80% margin but leaves customer success in operating expenses may really be at 65%. On a 15-month payback, that difference alone adds almost four months.

AI features make this more pressing. Model inference is a per-use cost that scales with customer activity (Snowflake's filing lists it inside cost of revenue), so heavy users of an AI feature can have a much lower margin than light users on the same plan.

LTV: cap the lifetime

The textbook formula is:

LTV = monthly revenue per account × gross margin ÷ monthly churn rate

Dividing by churn assumes customers leave at a steady rate forever. At 1% monthly churn, that implies an average lifetime of 100 months, over eight years. Few SaaS companies have enough history to support that, and a product built in 2026 may look very different by 2034. Many finance teams cap LTV at three to five years of gross profit, or use a discount rate, so a low churn estimate cannot inflate the ratio.

Why simple payback misleads

Simple payback = CAC ÷ (monthly revenue per account × gross margin)

It measures how long a customer who never churns takes to pay back. The average acquired customer does churn, so the real break-even point is later. The churn-adjusted version asks when cumulative expected gross profit per acquired customer reaches CAC. With monthly gross profit g and monthly churn c, expected cumulative gross profit after n months is:

g × (1 − (1 − c)ⁿ) ÷ c

That total can never exceed g ÷ c, which is the uncapped LTV. If CAC is bigger than that, the average customer never pays back at all.

A worked example

Take a product with a $6,000 CAC, $500 of monthly revenue per account, and an 80% gross margin, so each active customer produces $400 of gross profit a month. Simple payback is 15 months. Here is what churn does to it (figures are illustrative):

All three rows report the same 15-month simple payback. Only the churn-adjusted figure shows that the third business loses money on the average customer. If your board deck shows only simple payback, add the churn-adjusted number next to it.

Annual contracts change cash payback, not unit economics

When a customer prepays a year, you collect $6,000 up front in the example above instead of $500 a month. That shortens cash payback dramatically, which matters a lot for runway. It does not change gross profit per customer or LTV. Keep the two separate. And if you discount annual prepayment, use the discounted price in every formula.

Reading LTV:CAC in context

The widely quoted target of 3:1 is a rule of thumb, not a law. David Skok's SaaS guidelines say the best SaaS businesses run above 3:1 (sometimes 7 or 8) and many recover CAC in 5 to 7 months, and that profitability turns anemic once recovery takes longer than 12 months. He also says these are guidelines with exceptions. Here is a reasonable way to read the ratio:

The ratio only means something when it is split by segment. Enterprise customers often cost several times more to win than SMB customers and stay much longer. Blending them hides whether either segment works on its own.

Cohorts: the only way to see a trend

Averages over the whole customer base mix customers acquired three years ago in an easy market with customers acquired last month. A cohort table groups customers by the month or quarter they signed and follows each group separately. A simple version looks like this (figures are illustrative):

This table shows a business that is getting worse even while it grows: each cohort cost more to acquire and keeps less revenue. A blended LTV:CAC would still look fine for a year or more. Questions a cohort view answers that averages cannot:

• Is CAC rising as the best channels saturate?
• Does churn level off after the first 90 days, or keep going?
• Which acquisition channels produce customers who expand rather than churn?
• Did the last price change hurt retention in the cohorts that signed after it?

Company-level efficiency checks

Unit economics describe one customer. Investors also look at the whole company, using a few ratios built from the same data.

Net revenue retention (NRR). (Starting ARR from a cohort − churned ARR − contraction + expansion) ÷ starting ARR, usually over 12 months. Above 100% means existing customers grow on their own, which raises LTV without extra CAC. Definitions differ between companies. Datadog takes the ARR of customers from 12 months earlier, nets out contraction and attrition, and excludes new customers (about 120% at December 31, 2025). Snowflake compares a cohort's product revenue across a two-year window (125% at January 31, 2026). Check the definition before comparing your number with either. Our SaaS CLV guide covers how to raise it.

Magic number. Net new ARR in a quarter ÷ sales and marketing spend in the prior quarter. Scale Venture Partners traces the term to its 2005 look at Omniture, which generated more than $2 of first-year revenue per $1 of go-to-market spend. There is no universal cutoff, so track your own trend and compare it by segment.

Burn multiple. Net cash burn ÷ net new ARR over the same period. David Sacks of Craft Ventures introduced it in April 2020 as a whole-company check. His examples: burning $2M to add $1M of ARR is a 2x multiple, reasonable for an early-stage startup, while burning $5M for the same $1M is 5x and a reason to cut costs.

Rule of 40. Revenue growth rate plus profit margin (often free cash flow margin) should reach 40%. Brad Feld wrote up the rule in 2015 for SaaS companies at scale, meaning roughly $50 million or more in revenue, so it is a poor yardstick for an early-stage company. A company growing 50% with a −10% margin passes; so does one growing 15% with a 25% margin.

What to fix first

Payback has four inputs, so there are four levers.

1. Lower CAC. Shift budget toward channels with the lowest CAC and acceptable retention. Referral and partner channels are often cheaper, but check their cohorts before scaling.
2. Raise price or revenue per account. Packaging and pricing changes act on every new customer at once. Our SaaS pricing strategies guide covers value metrics and tier design.
3. Protect gross margin. Watch hosting and inference cost per account, and automate support tasks that scale with customer count.
4. Cut churn, especially early churn. In the table above, going from 4% to 2% monthly churn shortens real payback by about five months and doubles LTV. The churn reduction guide goes through onboarding and at-risk signals.

If payback is long but cohorts are sound, the problem may be financing rather than economics. Customers that pay back in 24 months are still profitable; you just need capital to carry them. Revenue-based financing and venture debt are built for that gap.

A monthly checklist

• Recalculate CAC by channel and segment, with spend lagged by your sales cycle.
• Compare assumed gross margin with actual cost of revenue, including support and inference.
• Report both simple and churn-adjusted payback for each segment.
• Update the cohort table and flag any cohort that retains less than the one before it.
• Track NRR and burn multiple alongside the per-customer numbers.

RevOps teams usually own this reporting, since it needs marketing, sales, finance, and product data in one place.

This article is for informational purposes only and is not financial or investment advice. Benchmarks vary by segment, market, and funding environment.