Unit Economics Deep Dive — LTV, CAC, and Payback
The LTV/CAC ratio is the single most important diagnostic of a venture-scale business — here is how to compute it honestly and what it actually tells you.
🎯 By the end you'll understand
How to compute fully-loaded CAC — including founder time at an imputed hourly cost — so it is honest rather than flattering. How LTV embeds three assumptions that make it a model, not a measurement. How payback period gives you operational insight that LTV/CAC alone cannot. How to use sensitivity analysis to identify the single lever that moves your unit economics the most. And how to recognize the common calculation that sounds rigorous but produces a useless number — the one investors can spot immediately and founders are reluctant to correct.
KEY INSIGHTLTV/CAC is a model, not a measurement — and the assumptions are the point
Every founder who has ever shown a pitch deck knows their LTV/CAC ratio. About half of them computed it in a way that would not survive a CFO's first pass. The ratio is not the insight — the embedded assumptions are.
At Forge — 18 customers, $780K ARR, NRR at 112% — the unit economics question has moved from theoretical to operational. You can compute LTV/CAC now with real data. The question is whether you compute it honestly.
The classic mistake is subtle. LTV/CAC of 6.2x sounds like a healthy B2B SaaS business. But the number was computed with a 5-year LTV window (most SaaS companies cannot reliably predict 5-year churn from 18 months of data), and with a CAC figure that counts only the marketing spend but not the founder hours spent in each deal cycle. Recomputed with a 3-year LTV window and founder time at a reasonable imputed rate, the same business comes in at 3.1x — acceptable but not exceptional, and a very different conversation with the team.
A well-computed LTV/CAC tells you three things: whether your business model works at current scale, what the binding constraint on profitability is (CAC efficiency, retention, or gross margin), and which lever, improved by 20%, has the biggest effect on the ratio. A badly computed one tells you nothing useful and gives investors a reason not to trust your other numbers.
CORE CONCEPTFully-loaded CAC: what actually counts
Customer Acquisition Cost is all spending attributable to acquiring a new paying customer, divided by the number of new customers acquired in the same period. The formula is simple. The devil is in what you include in the numerator.
| Cost category | What to include | Common omission |
|---|---|---|
| Sales team | Full compensation (salary + benefits + commissions + equity cost), tools (CRM, sales engagement), travel, entertainment | Equity cost is often excluded; benefits are sometimes omitted; entertainment gets classified as "business development" |
| Marketing spend | All paid media, content production costs, SEO tools, event sponsorships, lead-gen services, agency fees | Content production is often expensed separately; events are sometimes classified as brand rather than demand-gen |
| Founder time | Hours spent in sales calls, proposals, deal structuring, customer negotiation — imputed at a fair hourly rate | Almost always omitted entirely. This is the biggest distortion in early-stage unit economics. If the founder closes 80% of deals and spends 20 hours per deal, that is a real cost the business must eventually replace. |
| Onboarding / CS pre-revenue | CS labor in the trial and POC phase, before the customer signs, belongs in CAC not in customer success costs | Often classified as post-sale CS cost, depressing the apparent CAC while inflating apparent CS efficiency |
The imputed rate for founder time is contested. One reasonable approach: if you would hire a VP of Sales for $200K total comp to replace yourself in the sales process, and that VP's effective hourly cost is about $95/hour (at 2,080 hours per year), then your time is worth roughly that much to the business per hour spent in a deal. Some founders use a lower figure ($50–70/hour) for very early-stage work where the activity is also learning. The exact figure matters less than including it.
For each hour you spend in the sales process, ask: if I had a sales hire doing exactly this activity, what would it cost? Apply that rate to your hours. This makes the founder-time estimate explicit and defensible, and it surfaces the question every B2B founder needs to answer: can the sales motion be replicated by someone other than the founder, and at what cost?
CORE CONCEPTLTV: the model and its three embedded assumptions
Lifetime Value is the net present value of expected gross profit from a customer over their entire relationship with you. The standard simplified formula for SaaS is:
LTV = (ACV × Gross Margin %) ÷ Churn Rate
Where ACV is Annual Contract Value, Gross Margin % is revenue minus COGS divided by revenue, and Churn Rate is annual logo churn. Example: $43,333 ACV × 75% GM ÷ 18% churn = $180,554 LTV.
This formula embeds three assumptions that are worth naming explicitly, because getting them wrong produces an LTV that is technically computed but empirically meaningless.
Assumption: churn rate is stationary
The formula treats churn as a fixed rate applied indefinitely. In practice, cohort churn is typically front-loaded — customers who churn at month 6 represent a different population than customers who survive 24 months. Early-stage SaaS companies with fewer than 3 years of customer data cannot reliably estimate long-run churn from short-run data. A company with 8% monthly churn does NOT have annual churn of 96%; annualized monthly churn is 1 − (1 − 0.08)^12 = 63% — but even that arithmetic conceals the shape of the hazard function. Using your current churn rate projected to infinity overstates LTV when churn is falling with customer age (which it typically is).
Assumption: ACV is constant or grows at a fixed rate
The formula uses a single ACV figure. If your existing customers expand (NRR > 100%), LTV is understated; if they compress (discounts, downgrades), it is overstated. Forge's NRR of 112% means expansion revenue is real and significant — the LTV formula without expansion will understate true customer value by roughly the magnitude of expansion over the customer lifetime. A more honest LTV includes an expansion adjustment, but this requires estimating future expansion rates that you cannot observe from early data.
Assumption: gross margin is stable
Early-stage gross margin often changes significantly as you hire customer success, invest in infrastructure, or onboard customers at high service intensity. Using current gross margin to project LTV implicitly assumes that the cost structure supporting future customers looks like your cost structure today. For most early SaaS companies, this is aspirational — the current GM reflects founder-subsidized CS labor that will eventually need to be replaced.
The LTV formula computes an infinite series (or until churn reaches 100%). The practical convention is to cap LTV at a defined window — typically 3 or 5 years. A 5-year LTV window for a company with 18 months of customer data is extrapolating 40 months beyond observed behavior. The number is not wrong by construction — it is a model. But presenting it to an investor without disclosing the window and the data vintage is misleading. Use a 3-year LTV cap unless you have multi-year cohort data that justifies the extension.
HOW IT WORKSCAC payback period: the operational metric
CAC payback period is the number of months required to recover the cost of acquiring a customer from the gross margin generated by that customer. The formula:
Payback = CAC ÷ (MRR × Gross Margin %)
Where MRR is the monthly recurring revenue from the customer. Example: $22,000 CAC ÷ ($3,611 MRR × 75% GM) = 8.1 months payback.
Payback period is the operationally useful unit economics metric because it tells you something LTV/CAC does not: how much cash the business consumes per new customer before it starts generating a return. A company with LTV/CAC of 4x and 36-month payback is burning significantly more cash to grow than a company with LTV/CAC of 3x and 14-month payback. In a capital-constrained environment, the second company can grow faster.
| Payback period | Stage-appropriate benchmark | Implication |
|---|---|---|
| < 12 months | Strong for early-stage B2B SaaS | Business can grow without proportional capital investment; each customer funds the next |
| 12–24 months | Acceptable for enterprise B2B | Growth requires working capital; manageable with a modest fundraise |
| 24–36 months | Concerning for Series A stage | Capital-hungry growth; investor pressure to improve either CAC or GM before scaling |
| > 36 months | Difficult to fund at Series A | Unit economics need structural improvement before growth makes sense; growth compounds the problem |
The relationship between payback period and LTV/CAC depends heavily on churn. A low-churn business (5% annual churn) with 18-month payback has outstanding LTV/CAC because customers stay for a long time. A high-churn business (30% annual) with 18-month payback has mediocre LTV/CAC because customers leave before they generate meaningful return. This is why the payback period and LTV/CAC are complementary diagnostics, not substitutes — you need both.
SEE ITUnit Economics Dashboard — compute Forge's honest numbers
The sensitivity analysis reveals something counterintuitive: for most early B2B SaaS businesses at Forge's stage, churn reduction has a larger effect on LTV/CAC than CAC reduction of the same magnitude. This is because churn compounds — a 14% annual churn rate means you replace your entire customer base every 7 years; a 10% churn rate means every 10 years. Each point of churn reduced increases LTV by a factor that scales with the current churn rate. Meanwhile, reducing CAC by 20% requires either reducing founder time (which may not be possible) or improving sales efficiency — which is slow and hard. Retention is a faster lever.
FORGEReading Forge's unit economics at 18 customers
Forge's milestone data: 18 customers, $780K ARR ($43,333 ACV average), NRR at 112%, first CS hire made, founder exited day-to-day support, growth predictable within 15% month-over-month.
With those numbers and reasonable assumptions, here is what Forge's unit economics actually look like — and what they are telling you:
Fully-loaded CAC is probably $18,000–$25,000
Forge is closing roughly 2 customers per month. Monthly spend on sales (partial VP-equivalent salary plus marketing): ~$14,000. Founder time at this stage has not yet been fully replaced — even with the CS hire, the founder is likely still involved in 50–60% of deals for the first several months post-hire. At 10–15 founder hours per deal and $90/hr imputed, that is $900–$1,350 per deal. Pre-sale CS for the POC process: ~$1,000 per deal. Blended monthly total of about $15,300 ÷ 2 new customers = $7,650 base CAC plus founder time plus pre-sale CS = $9,550–$10,500. Wait — this seems low. But Forge is in the "founder still closing" phase where sales efficiency is partially subsidized by founder capital. As the founder exits and a full sales team is hired, the CAC will rise before it falls. The real question is not today's CAC — it is what CAC looks like at 5 new customers per month with a full sales team, no founder time.
LTV looks strong, but 18 months of data means the 3-year cap is mostly extrapolation
At $43,333 ACV, 75% gross margin, and 14% annual churn: LTV (3-year cap) = $43,333 × 0.75 × 3 = $97,500. With 12% seat-expansion rate, expansion revenue = $43,333 × 0.12 = $5,200/yr; expansion gross profit = $5,200 × 0.75 = $3,900/yr. Three-year expansion GP (compounding): $3,900 + $4,368 + $4,892 ≈ $13,160. Honest 3-year LTV with expansion: approximately $108,000–$111,000 depending on exact churn timing assumption.
The payback period is the real watch number at this stage
At fully-loaded CAC of ~$22,000 and MRR of $3,611 at 75% GM, payback = $22,000 ÷ ($3,611 × 0.75) = $22,000 ÷ $2,708 = 8.1 months. That is strong. What it tells you: Forge can grow without consuming proportional capital — each customer is recovered within the fiscal year, before the next customer acquisition budget is due. This is the data that supports raising a Series A to accelerate growth rather than to sustain operations. A 24+ month payback would change that conversation entirely.
The unit economics at Stage 5 are telling you two things. First: the business model works — LTV/CAC above 3x with payback under 12 months is a fundable profile. Second: the numbers will deteriorate before they improve as you replace founder-subsidized CAC with real sales team cost. A Series A investor who understands this will model a CAC increase in their first 12 months post-investment and ask whether the LTV numbers hold. Your answer: yes, if retention continues. The 112% NRR is your strongest unit economics asset right now.
WEAKNESSWhere this bites
The canonical failure mode: LTV computed with a 5-year window using 12 months of churn data, and CAC computed without founder time because "the founder's time doesn't count as a cost." At Forge's numbers, this produces LTV/CAC of roughly 8–10x — a number that sounds like a unicorn and impresses a first-time investor. A Series A investor with a CFO in the room will ask two questions: "What is your payback period?" and "What does CAC look like at 20 customers per month with a full sales team?" If you cannot answer either, they already know you did the flattering version of the calculation. Honesty in unit economics is not just ethical — it is strategically useful, because it forces you to know the real constraint before you scale into it.
LTV/CAC of 4x says nothing about whether you can afford to grow. A company with LTV/CAC of 4x and 30-month payback is burning 30 months of gross margin per new customer before it recovers the acquisition cost. At 3 new customers per month, that is $66,000 per month in unreturned CAC investment, building toward a peak cash requirement of roughly $1.98M (30 months × $66K) if growth is linear. A company with LTV/CAC of 3x and 9-month payback consumes far less capital to grow at the same pace. The payback period and the growth rate together determine cash consumption — LTV/CAC alone does not.
Your blended CAC is an average across all channels and deal sizes. The marginal cost of acquiring the next customer — especially if you are moving upmarket or into a new segment — may be significantly higher than the average. Early-stage companies often improve average CAC over time as they scale efficient channels, while simultaneously deteriorating the marginal CAC on harder-to-reach segments. Presenting average CAC to model future unit economics while simultaneously pursuing a higher-CAC expansion strategy overstates the quality of future economics. Track CAC by channel and by deal size whenever you have enough data to segment it.
KEY INSIGHTHow this connects
- 3.2 · Value-Based Pricing — The Only Pricing That Survives — ACV, the numerator in your LTV calculation, is the direct output of your pricing strategy. If you priced on cost-plus or competitive parity rather than value, you have systematically underpriced into a lower LTV. The lesson on value-based pricing is the upstream input to the unit economics calculation — revisit it if LTV/CAC is constrained by ACV rather than margin or churn.
- 3.6 · Gross Margin — The Number That Controls Your Destiny — Gross margin is embedded directly in both LTV and payback period. Every point of GM improvement flows through to both metrics multiplicatively. If your LTV/CAC is constrained by margin rather than CAC or churn, the gross margin lesson is the intervention point.
- 4.8 · Pricing in the Sales Process — Holding the Line and Negotiating Smartly — Discounting directly reduces realized ACV and therefore LTV. A CAC that closes a deal at 20% discount is not the same as a CAC that closes at full price. The unit economics calculation with discounted ACV gives you the unit economics you will actually experience, not the ones on your pricing page.
- 5.1 · Product-Market Fit — The Measurable Version — NRR, the expansion signal in your LTV model, is the quantitative output of product-market fit. If NRR is above 110%, expansion is contributing meaningfully to LTV. If NRR is below 100%, you have churn and contraction eating the LTV model faster than the simple churn rate suggests. The connection between PMF and unit economics runs through NRR.
- 5.5 · Cohort Analysis — Reading the Truth Behind Aggregate Metrics — The single most important validation of your LTV model is whether cohort data actually supports the churn rate assumption you embedded in the calculation. Lesson 5.5 shows how to build the cohort waterfall that either confirms or challenges the LTV number you computed here.
GLOSSARYKey terms
- Customer Acquisition Cost (CAC)
- All costs attributable to acquiring a new paying customer, divided by the number of new customers acquired in the same period. Fully-loaded CAC includes sales team, marketing spend, tools, events, founder time at an imputed rate, and pre-sale customer success costs.
- Lifetime Value (LTV)
- The net present value of expected gross profit from a customer over their full relationship with the company. In simplified SaaS form: (ACV × Gross Margin %) ÷ Annual Churn Rate, typically capped at a defined window (3 or 5 years). LTV is a model with embedded assumptions about churn stationarity, ACV stability, and gross margin — not a measurement.
- LTV/CAC ratio
- The ratio of lifetime value to customer acquisition cost. The widely cited benchmark for venture-scale B2B SaaS is ≥ 3x. A ratio above 3x suggests the business creates more value per customer than it costs to acquire them; below 3x suggests a structurally challenged unit economics profile that either limits growth or requires a capital subsidy to sustain.
- CAC payback period
- The number of months required to recover the CAC from gross margin generated by the customer. Formula: CAC ÷ (MRR × Gross Margin %). A metric of capital efficiency that LTV/CAC does not capture — two companies can have identical LTV/CAC ratios with very different payback periods and therefore very different capital requirements for growth.
- Imputed founder cost
- The cost assigned to founder time spent in revenue-generating activities (sales, deal structuring, negotiation) for the purpose of computing fully-loaded CAC. Typically derived from what it would cost to hire a sales role to replace the founder's function. Omitting it produces a CAC that understates the true cost of the current sales motion and overstates what unit economics will look like with a real sales team.
- Gross margin contribution
- The gross profit generated per customer per period, used in the denominator of the payback period calculation. ACV × Gross Margin % ÷ 12 = monthly gross margin contribution per customer. This is the rate at which a customer is "paying back" the CAC.
- Churn stationarity assumption
- The implicit assumption in the standard LTV formula that the churn rate observed today will remain constant throughout the customer's lifetime. In practice, churn is typically front-loaded — newer cohorts churn faster than mature ones. This assumption makes the formula a model rather than a projection, and means that early-stage LTV calculations extrapolate beyond observed behavior.
ACTIVE RECALLCheck your understanding
Answer in your own words first, then reveal.
1A founder says their CAC is $8,400. They calculated it as total marketing spend divided by new customers. What is likely missing, and by how much might the fully-loaded CAC differ?›
The calculation almost certainly omits: sales team compensation (salary, commissions, benefits, equity cost); sales tools (CRM, sales engagement platform, prospecting tools); events and travel; founder time — the largest single omission in early-stage companies, often worth $5,000–$20,000 per deal at an imputed rate of $80–$100/hour over 15–25 hours per deal; and pre-sale customer success costs (POC and trial support before contract signature). In a typical early-stage B2B SaaS company, the fully-loaded CAC is often 1.5–2.5× the marketing-only CAC. So $8,400 in marketing spend may correspond to a fully-loaded CAC of $12,600–$21,000. The specific delta depends on how founder-dependent the sales process is — the higher the founder involvement, the larger the gap.
2Why does the standard LTV formula (ACV × GM ÷ churn) embed an assumption that makes it a model rather than a measurement? What are the three embedded assumptions?›
The three embedded assumptions are: (1) Churn stationarity — the formula assumes the churn rate observed today will remain constant for the customer's entire lifetime, which ignores the empirical reality that churn is typically front-loaded and falls as customers mature. This means early-stage LTV calculations extrapolate beyond observed data into a regime with lower churn than you have actually seen. (2) ACV stability — the formula uses a single ACV figure and implicitly holds it constant. If NRR > 100%, the model understates LTV; if customers discount or downgrade, it overstates it. (3) Gross margin stability — the formula assumes current GM holds throughout the customer's lifetime, ignoring the common early-stage reality that founder-subsidized CS labor depresses apparent costs and that a mature cost structure (with a real CS team) will have different unit economics. Together, these three assumptions mean LTV is a model of a specific scenario, not a forecast of what will happen.
3Two companies both have LTV/CAC of 4x. Company A has a 10-month payback period; Company B has a 30-month payback period. Which is in a better position to grow without external capital, and why?›
Company A is in a significantly better position. The LTV/CAC ratio describes the ultimate return on customer acquisition but says nothing about when that return arrives. Company A recovers its customer acquisition cost within 10 months — meaning at steady state, each customer is funding its own replacement within the fiscal year, and growth is largely self-financing. Company B takes 30 months to recover CAC: at any given time, it has 30 months of unrecovered CAC "in the air" for each active customer. If Company B grows at 5 new customers per month, it is carrying $150K × 5 = potentially $750K+ in unreturned CAC investment at any given time, building toward a cash requirement that scales directly with growth rate. Company B needs external capital to fund the gap between CAC payment and revenue recovery — Company A needs much less. This is why payback period is the key capital-efficiency metric, and why two companies with identical LTV/CAC ratios can have radically different funding requirements.
4The sensitivity analysis in the widget shows that reducing churn by 20% produces a larger LTV/CAC gain than reducing CAC by 20%. Why is this the case for most early B2B SaaS businesses?›
Because LTV = (ACV × GM) ÷ churn rate — churn appears in the denominator. When churn is the binding constraint, a 20% reduction in churn rate (e.g., from 15% to 12%) increases LTV by a factor of 15/12 = 1.25, or 25%. Meanwhile, a 20% reduction in CAC reduces the denominator of LTV/CAC by 20%, producing an identical 25% improvement in the ratio only if all else is equal. However, the mechanism differs: churn improvement is multiplicative on the LTV side (increases both LTV and the effective customer lifetime), while CAC reduction is linear on the cost side. More importantly, at typical early-stage churn levels (10–25% annual), churn reduction is the faster operational lever — it requires retaining existing customers, which the CS hire and expanded onboarding process can directly address. Reducing CAC requires improving sales efficiency, which is slower and harder. The sensitivity analysis shows that working on what you already have (retention) is usually better than making new customer acquisition cheaper.