Unit Economics for SaaS: The Metrics That Matter

Unit economics for SaaS comes down to a short list of recurring-revenue metrics that show whether each customer you sign is worth more than it costs to win and serve, and how fast that value comes back. Subscription software has predictable revenue and high gross margins, which makes its unit economics look easy, but small errors in churn, margin or acquisition cost compound over years. This guide defines the core metrics, works every one of them through a single hypothetical company with checked arithmetic, and explains how to read them together instead of one at a time. It belongs to unit economics and budgeting and forecasting, follows unit economics for ecommerce, and leads to unit economics for marketplaces and common unit economics mistakes.

The unit in SaaS

The natural unit is the customer account, measured per month. The central questions are:

1. How much recurring revenue does an account bring in each month?
2. How much of that is left after the cost of serving the account?
3. How long does the account stay?
4. How much does it cost to acquire the account, and how quickly does that cost come back?
5. Does the account grow or shrink while it stays?

Each metric below answers one of those questions.

The worked company

A hypothetical B2B software company (illustrative figures):

Metric 1: MRR and ARR

Monthly recurring revenue (MRR) is the normalized monthly value of all active subscriptions. Annual recurring revenue (ARR) is MRR × 12.

MRR = 200 accounts × $500 = $100,000. ARR = $100,000 × 12 = $1,200,000.

Only count recurring revenue. One-time setup fees and usage overages that fluctuate belong elsewhere. See MRR vs. ARR and annual recurring revenue.

Metric 2: Subscription gross margin

Gross margin for SaaS includes hosting, third-party software that powers the product, payment fees, customer support and customer success costs that serve existing accounts. Gross margin of 80 percent means that of $500 per account, $400 remains after the cost of serving. Be careful about what is placed in cost of revenue, because software companies differ, and a comparison between two companies is only valid if the definitions match. See SaaS gross margin.

Metric 3: Churn and lifetime

Monthly logo churn is the share of accounts lost each month. At 2 percent, expected lifetime is 1 / 0.02 = 50 months. In the example, the company loses about 2% × 200 = 4 accounts a month, or 12 a quarter. See churn rate and revenue churn vs. logo churn.

Churn is not constant for most products. New accounts often churn faster than mature ones, and small accounts churn faster than large ones. Use cohort analysis to check the assumption before relying on the 50-month figure. See cohort analysis.

Metric 4: Customer lifetime value (LTV)

LTV on a gross-margin basis = ARPA × gross margin × lifetime = $500 × 0.80 × 50 = $20,000.

Using revenue instead of gross margin would give $25,000 and overstate the value by 25 percent. Always use margin. See customer lifetime value.

Metric 5: CAC

CAC = sales and marketing spend / new customers = $150,000 / 30 = $5,000. Include salaries, commissions, tools, program spend and an allocation of overhead for the sales and marketing team. See customer acquisition cost.

Metric 6: LTV to CAC and payback

LTV : CAC = $20,000 / $5,000 = 4.0.

CAC payback in months = CAC / (ARPA × gross margin) = $5,000 / ($500 × 0.80) = $5,000 / $400 = 12.5 months.

A common rule of thumb treats an LTV to CAC ratio of about 3 or higher as healthy, and payback of a year or less as good for small accounts. These are conventions, not laws. They depend on growth stage, churn assumptions and available capital. A 4.0 ratio with a 12.5-month payback means the company needs to finance a year of each cohort's costs before the cash returns. See LTV to CAC ratio and CAC payback period.

Metric 7: Net new MRR

Net new MRR combines the four movements in recurring revenue during a period. In the quarter:

Check: 15,000 + 3,000 - 1,000 - 6,000 = 11,000. Starting MRR was $89,000 and ending MRR is $100,000, which matches.

Metric 8: Net and gross revenue retention

Retention is measured on a cohort of customers over a period, usually twelve months. Suppose the customers who existed twelve months ago produced $70,000 of MRR then.

• Gross revenue retention (GRR): MRR from that cohort today, counting only churn and contraction, with expansion excluded. If it is $62,300, GRR = 62,300 / 70,000 = 89%.
• Net revenue retention (NRR): MRR from the same cohort today, including expansion. If it is $72,100, NRR = 72,100 / 70,000 = 103%.

NRR above 100 percent means the existing base grows even with no new customers, which lowers the pressure on new sales. GRR shows how much of the base survives before expansion rescues it. Read them together: an NRR of 103 percent built on a GRR of 89 percent depends on a steady flow of upsells to cover leakage. See net revenue retention and gross revenue retention.

Metric 9: Efficiency of growth spending

Two ratios show whether spending converts to growth.

Net new ARR per dollar of sales and marketing. Net new MRR of $11,000 is $132,000 of ARR (11,000 × 12). Divided by $150,000 of spend, the quarter returns 0.88. This is a simplified version of the SaaS magic number. See the SaaS magic number.

Burn multiple. Net burn divided by net new ARR: $300,000 / $132,000 = 2.3. Each dollar of net new ARR cost $2.30 of cash. Lower is better, and what is acceptable depends on stage and the cost of capital. See burn multiple.

Add the Rule of 40, the sum of growth rate and profit margin, if you want one summary figure. A company growing 40 percent with a free cash flow margin of minus 10 percent scores 30. See the Rule of 40.

Reading the metrics together

No single metric tells the story. Use them in combination:

Segment before you conclude

Averages hide the real picture in SaaS. Break the metrics down by:

• Customer size or plan. Small and large accounts have different churn and different CAC.
• Acquisition channel. Inbound, outbound, partner and self-serve differ in cost and quality.
• Cohort. Compare the first year of the 2025 cohort with the first year of the 2026 cohort.
• Product line. One line may subsidize another.

For example, if small accounts have ARPA of $150, gross margin 75 percent, churn 4 percent monthly and CAC $900, then payback = 900 / (150 × 0.75) = 8 months and lifetime = 25 months, so LTV = 150 × 0.75 × 25 = $2,812.50 and LTV : CAC = 3.1. If large accounts have ARPA $2,000, margin 82 percent, churn 1 percent and CAC $14,000, payback = 14,000 / (2,000 × 0.82) = 8.5 months, lifetime = 100 months, LTV = 2,000 × 0.82 × 100 = $164,000 and LTV : CAC = 11.7. The blended company number conceals a large difference in the quality of the two segments. A 100-month lifetime is an extrapolation, though, and a careful team would cap the lifetime or discount future cash flows.

How to build the model

1. Set up a monthly cohort table: month of acquisition by month since acquisition.
2. Track logos and MRR separately.
3. Record expansion and contraction as separate lines.
4. Calculate gross margin from cost of revenue that includes support and hosting.
5. Allocate sales and marketing spend to the period in which it produces customers, allowing for the sales cycle.
6. Compute LTV, CAC, payback, NRR and GRR for each segment.
7. Add a sensitivity table for churn, CAC and gross margin. See sensitivity analysis on unit economics.
8. Present the numbers as a dashboard. See how to build a unit economics dashboard.

Common mistakes

• Computing LTV from revenue, not gross margin.
• Using a lifetime formula with limited data. A company with 14 months of history should not claim a 60-month lifetime without caveats.
• Mixing monthly and annual churn.
• Ignoring expansion and contraction.
• Leaving customer success and support out of cost of revenue.
• Counting annual prepayments as MRR, or counting usage spikes as recurring.
• Using blended CAC when paid and outbound channels carry the marginal cost.
• Not stress-testing churn and CAC assumptions.

FAQs

What are the key unit economics metrics for SaaS?

MRR, ARPA, gross margin, churn, LTV, CAC, LTV to CAC ratio, CAC payback, net and gross revenue retention, and burn multiple.

How do you calculate LTV for SaaS?

Multiply ARPA by gross margin by expected lifetime in months, where lifetime is one divided by monthly churn. Check the result against cohort data.

What is a good CAC payback for SaaS?

It depends on account size and funding. Shorter is easier to finance, and many teams aim for a year or less for smaller accounts, though larger accounts may justify longer payback.

What is the difference between NRR and GRR?

NRR includes expansion revenue from the same customers, while GRR counts only churn and contraction and so cannot exceed 100 percent.

Do SaaS unit economics apply to hardware or consumer subscription businesses?

The framework applies to any recurring model, but the cost of goods, shipping and returns must be included, so the gross margin line looks different.