Unit Economics Wiki
LTV
LTV:CAC compares a customer's expected discounted gross profit with the cost of acquiring that same type of customer — and is uninterpretable unless both sides are defined.
Snapshot
What it is
LTV:CAC = contribution-based customer lifetime value ÷ customer acquisition cost, computed on the same customer unit, cohort, channel, and cost boundary.
Why it matters
It is the one number that forces acquisition and retention into a single model, and the one number investors will quote back at you. Which is exactly why it needs its definitions attached.
What it is not
It is not a standard, not audited, and not a law. The 3:1 benchmark is a venture-capital rule of thumb from a 2012 blog post, not an empirical finding — its author described the underlying figures as "early guesses" validated informally against private portfolio companies, and explicitly called them "only guidelines." No published study establishes 3.0 as a threshold of viability.
Key takeaways
One cohort honestly produces 2.08x to 6.67x depending on which of six standard LTV methods you pick. The ratio without the method is close to information-free.
Match both sides. Gross-profit LTV over fully loaded CAC is a different metric from revenue LTV over paid-media CAC, and the second looks roughly 3.2x better for free.
The ratio has no time axis. Two cohorts at 3.0x can pay back in 12 months and 30 months. Always report payback beside it.
Sampling error alone spans the benchmark. A 100-account cohort with one year of history supports a 95% range of roughly 2.2x–3.8x. You cannot measure your way to "we are above 3."
On this page11 sections
What is the LTV:CAC ratio?#
LTV:CAC = present value of expected future customer gross profit
─────────────────────────────────────────────────────
attributable acquisition cost per customer
The numerator is a forecast; the denominator is an allocation. Neither is an observation, and neither is standardised. That is the whole reason the ratio needs a disclosed method rather than a headline.
A workable reporting set is three ratios, not one:
| Ratio | Numerator | Denominator | Use it for |
|---|---|---|---|
Gross-profit LTV : fully loaded CAC | Discounted gross profit, explicit horizon | All acquisition-oriented S&M | Tracking over time; talking to investors |
Contribution LTV : fully loaded CAC | Also deducts customer-specific success, support, payment, renewal cost | All acquisition-oriented S&M | Judging whether the business model works |
Incremental LTV : incremental CAC | Value of the customers the next dollar buys | Experiment-backed marginal spend | A specific budget decision |
The first travels best across periods. The second is the economically complete one. The third is the only one that answers "should we spend more?" — and it requires an incrementality test, not an attribution model.
Where does the 3:1 benchmark come from, and is it real?#
It comes from David Skok's SaaS Metrics 2.0, published on the For Entrepreneurs blog while he was a partner at Matrix Partners. The relevant passage is worth reading in the original:
"Over the last two years, I have had the chance to validate these guidelines with many SaaS businesses, and it turns out that these early guesses have held up well. The best SaaS businesses have a LTV to CAC ratio that is higher than 3, sometimes as high as 7 or 8."
Three things are true about that sentence, and founders should hold all three:
- It is a genuine, useful practitioner observation. Skok was looking at real portfolio and peer data, and the pattern he describes — good businesses cluster above 3, great ones well above — is broadly consistent with what public SaaS companies look like.
- It is not an empirical result. There is no sample, no methodology, no control for survivorship, and no published dataset. Skok's own words are "early guesses" that "held up well," and he adds directly: "I should stress that these are only guidelines, there are always situations where it makes sense to break them." Nothing in the source claims 3.0 is a viability threshold.
- It was derived from mature companies with stable churn, and is routinely applied to seed-stage companies with eighteen months of history. That transfer is where the damage happens.
Then there is the arithmetic problem, which is worse than the provenance problem. The benchmark does not specify a method, and the method moves the answer by more than 3x. A company can move from 2.1x to 6.7x with no change in its business at all — just by switching from an explicit discounted five-year gross-profit LTV to an undiscounted revenue perpetuity. A benchmark that can be cleared by choosing a formula is not a test.
How to use it anyway. Treat 3:1 as a conversational shorthand, not a hurdle rate. Then say what you actually mean:
| Instead of | Say |
|---|---|
"Our LTV:CAC is 3.2x" | "Discounted five-year gross-profit LTV over fully loaded CAC is 2.1x; the undiscounted perpetuity version is 5.0x" |
"We beat the 3x benchmark" | "Gross-profit payback is 16 months and we are cash-flow positive on a cohort by month 17" |
"LTV:CAC proves the model works" | "Here is the matured-cohort back-test: predicted versus realised cumulative gross profit by cohort age" |
How do you calculate a defensible ratio?#
1. Match the customer unit#
Account LTV with account CAC; location with location; household with household. Comparing account-level acquisition spend to seat-level lifetime value is the fastest way to a wrong answer that survives review.
2. Match the cohort and channel#
The LTV forecast has to describe the customers the CAC denominator actually bought. If channel mix shifted this quarter, last year's company-wide retention curve does not describe this quarter's cohort.
3. Match the cost boundary — and state it#
Write the label out in full: "discounted five-year gross-profit LTV ÷ fully loaded CAC." If the label does not fit in the chart title, put it in the footnote, but do not drop it.
4. Report payback alongside, always#
gross-profit payback (months) = CAC / monthly gross profit per acquired customer
The ratio is silent on timing, and timing is what kills companies. See CAC Payback Period.
5. Stress the fragile inputs#
Vary retention, gross margin, expansion, discount rate, and CAC. Publish base, downside, and observed-to-date. A single point estimate implies a precision the data does not contain.
6. Back-test matured cohorts#
For every cohort old enough to check, compare predicted with realised cumulative gross profit and attribute the error. This is the only evidence that turns a forecast into a claim.
Key Facts
The 3:1 benchmark is a blog-post guideline, self-described as guesswork
Its author writes that the figures were "early guesses" later validated informally, that the best SaaS businesses run "higher than 3, sometimes as high as 7 or 8," and that "these are only guidelines." No sample size, methodology, or dataset is published.
Skok, *SaaS Metrics 2.0*, For EntrepreneursThe same segment can differ 3.3x by channel
HubSpot measured an LTV:CAC of 1.5 selling direct into the very-small-business market versus 5.0 selling through value-added resellers — and restructured from 12 direct/4 channel reps to 2 direct/25 channel within a year. The company average would have hidden both numbers.
Skok, *SaaS Metrics 2.0*Improving the numerator beats cutting the denominator by two orders of magnitude
Gupta, Lehmann and Stuart estimate a retention elasticity of customer value of 3–7 against an acquisition-cost elasticity of 0.02–0.32. A 1% retention gain is worth roughly 10–100x a 1% CAC reduction.
Gupta, Lehmann & Stuart, *Valuing Customers*, JMR 2004A healthy public company can carry a payback the ratio would never reveal
Similarweb reported an average CAC payback of 21 to 22 months as of Q2 2025 while generating a 58–62% contribution margin on its recurring customer base — a perfectly viable business whose cash exposure a lifetime ratio does not describe.
Similarweb Q2 2025 shareholder letterWorked example: one cohort, six ratios#
Use the shared cohort from the CAC and LTV pages: 30 accounts, blended ARPA $1,000 MRR, 75% gross margin ($750/month, $9,000/year gross profit), 85% annual retention, 12% discount rate, fully loaded CAC $12,000.
| LTV method | LTV | LTV:CAC | Verdict against "3:1" |
|---|---|---|---|
Revenue-based perpetuity | $80,000 | 6.67x | Passes spectacularly — and is wrong |
Gross-margin perpetuity, annual basis | $60,000 | 5.00x | Passes |
Gross-margin perpetuity, monthly basis | $55,754 | 4.65x | Passes |
Discounted perpetuity, 12% | $33,333 | 2.78x | Fails, narrowly |
Discounted perpetuity, Gupta convention | $28,333 | 2.36x | Fails |
Explicit five-year discounted | $24,941 | 2.08x | Fails |
No arithmetic error appears anywhere in that table. The business is identical in every row. The spread is 3.2x, entirely from method selection — which is the practical case against treating 3.0 as a threshold.
Now change the denominator too. Suppose the founder quotes paid-media CAC of $4,000 instead of the fully loaded $12,000, against the revenue perpetuity:
$80,000 / $4,000 = 20.0x
versus the defensible figure of 2.08x. A tenfold difference, produced entirely by unmatched definitions, with every individual number technically true.
And the ratio still hides the decision. Split the same cohort by channel, using the explicit five-year discounted gross-profit LTV:
| Channel | Accounts | CAC | Monthly gross profit | 5-yr discounted LTV | LTV:CAC | Gross-profit payback |
|---|---|---|---|---|---|---|
Inbound | 18 | $8,000 | $450 | $14,965 | 1.87x | 17.8 months |
Outbound | 12 | $18,000 | $1,200 | $39,906 | 2.22x | 15.0 months |
Blended | 30 | $12,000 | $750 | $24,941 | 2.08x | 16.0 months |
The cheap channel is the worse channel on both the ratio and the clock. A founder optimising for a low CAC would have scaled inbound and starved outbound — the exact inversion HubSpot found and corrected.
How much confidence does the ratio actually carry?#
Retention is estimated from a finite cohort over a short window, and the ratio inherits that error non-linearly because churn sits in a denominator. Observing 100 accounts for twelve months, 15 of which churn, gives r = 0.85 with a standard error of 0.0357 — a 95% interval of 78.0% to 92.0%:
| Cohort observed | 95% CI on retention | Implied LTV:CAC (discounted perpetuity) |
|---|---|---|
50 accounts | 75.1% – 94.9% | 2.03x – 4.39x |
100 accounts | 78.0% – 92.0% | 2.21x – 3.75x |
400 accounts | 81.5% – 88.5% | 2.46x – 3.19x |
Every one of those intervals straddles 3.0. Even at 400 accounts you cannot statistically distinguish "below benchmark" from "above benchmark" — and this table prices only sampling error, holding the constant-hazard model fixed. Model error, which is larger, is on top.
The correct output of an honest LTV:CAC exercise is therefore an interval and a method, not a number. Anyone presenting 3.2x from an eighteen-month-old company is reporting a modelling choice.
What are the common mistakes?#
- Unmatched definitions. Revenue LTV over paid-media CAC. Looks like 20x, means nothing.
- Treating 3.0 as a hurdle. It is a practitioner guideline about mature companies, published without a dataset, and its own author calls it a guideline.
- Ignoring payback. A 4x ratio that arrives over eight years is unfinanceable for a company with nine months of runway.
- Applying mature retention to a new channel. New cohorts have different fit, onboarding, and churn shape by construction.
- Reading a high ratio as good news. A very high ratio usually means underinvestment, a capacity constraint, or an overstated LTV — not excellence. If you are genuinely at 8x, the question is why you are not spending more.
When does the ratio break?#
Before you have matured cohorts. With no cohort older than the payback period, the numerator is entirely model and the ratio is a restatement of your assumptions.
It does not fund fixed costs. A 4x contribution LTV:CAC coexists comfortably with heavy losses when cohort volume is small or R&D and G&A are large. The ratio is about unit economics; solvency is about the whole P&L. Pair it with burn and runway and the Rule of 40.
It does not prove incrementality. Attributed customers may have converted without the spend. For a marginal budget decision, use experiment-backed incremental CAC or accept that you are guessing.
Marketplaces and two-sided models. Acquiring one side creates value on the other. Side-specific CAC against cross-side contribution has to be modelled jointly. See two-sided markets.
Expansion-driven optimism. LTV rises mechanically with an assumed expansion rate. If net revenue retention above 100% is fed into a perpetuity as the retention term, the formula returns an infinite or negative value — which is the source of most implausible ratios in circulation.
Frequently asked questions
01Is 3:1 a real benchmark?
It is a real and widely used heuristic, and it is not an empirical finding. It originates in a 2012–2013 practitioner blog post whose author describes the numbers as early guesses validated informally against private companies, and who explicitly calls them guidelines rather than rules. Treat it as a conversational starting point, and report your method and your payback instead of your position relative to 3.
02What ratio should I actually target?
Target a payback period your balance sheet can finance and a back-tested LTV model, then let the ratio be whatever it is. If you need a shape: a discounted gross-profit LTV comfortably above CAC, with payback inside 12–18 months for SMB and inside 24 for enterprise, is a business that can fund its own growth. The ratio is the output, not the goal.
03Why does my ratio look worse than my competitors'?
Usually because they are quoting a different metric. Ask which LTV method, which discount rate, which horizon, and which CAC boundary. In the example on this page, the same cohort yields 2.08x and 20.0x depending on those four answers.
04Can the ratio be too high?
Yes, and it is a common finding in companies that are underspending. A ratio above roughly 5x usually means you are leaving growth on the table, are capacity-constrained in sales, or have an LTV model that has never been back-tested. All three deserve investigation.
05Should I use LTV:CAC or payback?
Payback, first. It uses only observed near-term gross profit, needs no discount rate or terminal assumption, and answers the question that actually determines survival — how long the company must finance the gap. Use LTV:CAC as the second number, for the value that arrives after recovery.
Related concepts#
- Customer Lifetime Value — build the numerator from survival-weighted gross profit.
- Customer Acquisition Cost — define the denominator's cost boundary.
- CAC Payback Period — add the time axis the ratio lacks.
- Gross Margin — the delivery-cost boundary both sides depend on.
- Churn Rate — the input the ratio is most sensitive to.
- Cohort Analysis — keep segment, channel, and maturity comparisons valid.
- Burn Rate and Runway — whether you can finance the gap the ratio ignores.
- Rule of 40 — the company-level counterpart to a unit-level test.
Sources#
- Skok, D., SaaS Metrics 2.0 — A Guide to Measuring and Improving What Matters, For Entrepreneurs. Origin of the LTV:CAC > 3 guideline, the author's characterisation of the figures as "early guesses" and "only guidelines," and the HubSpot direct-versus-channel comparison (1.5 versus 5.0).
- Gupta, S., Lehmann, D. R., & Stuart, J. A., Valuing Customers, Journal of Marketing Research 41(1), 7–18, 2004. Retention elasticity of 3–7 versus acquisition-cost elasticity of 0.02–0.32, and the discounted customer-value formulation used in the worked example.
- Similarweb Ltd., Q2 2025 Shareholder Letter (Form 6-K, Exhibit 99.2), August 12, 2025. Average CAC payback of 21–22 months and a 58–62% contribution margin on the recurring customer base.
- Similarweb Ltd., First Quarter 2026 Results (Form 6-K, Exhibit 99.1), May 13, 2026. Company-published definitions of CAC, customer retention cost, and CAC payback period as supplemental operating metrics.
- US Securities and Exchange Commission, Non-GAAP Financial Measures — Compliance and Disclosure Interpretations, Division of Corporation Finance. Staff position on defining and consistently applying supplemental measures presented to investors.
Source-use note: LTV:CAC is not defined by any accounting standard and no standard-setter publishes a threshold for it. The 3:1 figure is attributable to a single practitioner source, cited above; readers should not treat it as an audited or empirically established benchmark. All worked figures on this page are illustrative constructions.
Note: This page is educational and does not constitute accounting, financial, or investment advice. LTV:CAC is an unaudited operating ratio built from a forecast and an allocation, both chosen by the preparer. Consult a qualified accountant or financial adviser before relying on it in investor materials, a valuation, or a financing document.
Author
Dr. Sarah Zou
Independent economist · EconNova
Commercial strategy for technical products, with a focus on pricing, unit economics, and the operating choices behind the model.
About SarahTopics
Cite this page
Suggested citation
Zou, S. (2026). LTV:CAC Ratio: A Consistent Test of Growth Economics. In Unit Economics. Pricing & Monetization Wiki. https://sarahzou.com/wiki/unit-economics/ltv-cac-ratio
Open license
Reuse with attribution
This content is available for reuse. When referencing or republishing it, please credit Dr. Sarah Zou and link back to the original source.
Licensed under Creative Commons Attribution 4.0 International. You may share and adapt the material with appropriate credit.