You’ve been running your SaaS business for 12 months - you have paying customers, some churn, and an investor who wants to see your unit economics. The challenge is real: your dataset is too small for statistical confidence, but too important to ignore.

Most advice on unit economics assumes you have years of data and hundreds of customers. When you have 50 customers and 8 churned, the standard formulas produce estimates with false precision. LTV = ARPU / Monthly Churn gives you a single number that implies certainty you don’t have.

Here’s how to build credible unit economics from limited data — honest about limitations but rigorous in methodology — in a way that investors respect.

LTV with Limited Data

The standard LTV formula (average revenue per customer divided by monthly churn rate) has a fundamental problem when applied to small datasets: it assumes your churn rate is stable and representative. With 12 months of data and a handful of churned customers, you don’t know if your churn rate has stabilised. A single additional churn event next month could change the calculation by 20%.

A more honest approach: calculate observed LTV from your actual cohort data. Look at the customers who signed up in your first month. What’s the total revenue they’ve generated to date? What about Month 2’s cohort? Month 3’s?

This gives you observed LTV at different maturity points. Your Month 1 cohort has 12 months of data; your Month 6 cohort has 6 months. You can see whether the retention curve is flattening (good) or continuing to decline (concerning).

Present LTV as a range rather than a point estimate. For example: “Based on 12 months of cohort data, observed LTV is between $4,200 and $5,800, with the range narrowing as older cohorts stabilise.” This is more intellectually honest than “LTV is $5,100” and investors appreciate the transparency.

If your retention curves haven’t stabilised after 6 months, say so. “Insufficient data for confident LTV” is a better answer than a number you can’t defend.

CAC Across Channels

Blended CAC is useful as a headline number, but investors will always ask about channel-specific CAC. This requires discipline in tracking and attribution.

Start by listing every acquisition channel: outbound sales, inbound marketing, paid advertising, referrals, partnerships, and founder network (personal connections, warm introductions, conference conversations). Then allocate all acquisition costs to those channels.

The tricky part is founder time. If you spent 60 hours last month in sales activities, that has a cost. Not accounting for it dramatically understates your true CAC. A reasonable approach: value founder sales time at the market rate for a sales professional in your segment, and include it in the channel-specific CAC for whatever channel those hours supported.

The critical distinction investors look for: which channels are scalable, and which aren’t? Founder network is not scalable. Conference serendipity is not scalable. Outbound with a repeatable playbook is scalable. Content marketing with growing organic traffic is scalable. Your model should show the transition from non-scalable early channels to scalable growth channels, with CAC assumptions that reflect the difference.

Payback Period

Payback period answers a simple operational question: how long until a customer has generated enough gross margin to cover the cost of acquiring them?

The calculation: CAC divided by (monthly ARPU multiplied by gross margin percentage). If your CAC is $3,000, your ARPU is $800, and your gross margin is 75%, your payback period is $3,000 / ($800 × 0.75) = 5 months.

At early stage, payback period matters more than LTV for a practical reason: it tells you how quickly your customer acquisition spending recycles. If your payback is 5 months, every dollar you spend on acquisition comes back in 5 months and can be spent again. If your payback is 18 months, you need significant capital to fund growth because you’re waiting a long time to recoup each acquisition investment.

Generally, payback under 12 months is strong for B2B SaaS. Under 6 months is excellent. Above 15 months raises questions about whether the economics work without significant scale advantages.

Cohort Analysis Basics

A cohort retention table tracks how each month’s new customers retain over time. It’s the single most valuable analytical tool for a SaaS business, and it’s simpler to build than most founders think.

Structure it as a grid. Rows are cohorts (Month 1 customers, Month 2 customers, etc.). Columns are periods since acquisition (Month 0, Month 1, Month 2, etc.). Each cell shows the percentage of the original cohort still active.

What you’re looking for is the shape of the retention curve. A healthy SaaS business shows curves that decline initially and then flatten. The point where they flatten indicates your natural retention rate. If your Month 1 cohort retained 95% after Month 1, 88% after Month 2, 83% after Month 3, and 81% after Months 4 through 8, you have a flattening curve at around 80% — meaning you’ll retain roughly 80% of any cohort indefinitely after the initial churn period.

Concerning patterns: curves that continue to decline without flattening, or more recent cohorts that retain worse than older ones (suggesting product-market fit may be weakening). Both of these require honest acknowledgment and a plan to address.

The Metrics Investors Want to See

Here’s a concise checklist of what seed and Series A investors expect, with benchmarks:

LTV:CAC Ratio (Lifetime Value to Customer Acquisition Cost)

  • Good Range: 3:1 to 5:1

  • 🚩 Red Flag: Below 3:1 (unsustainable) or above 10:1 (underinvesting)

Measure your current LTV:CAC ratio with our free calculator.

CAC Payback Period

  • Good Range: 6–12 months

  • 🚩 Red Flag: Above 15 months

Gross Margin

  • Good Range: 70–85% (software)

  • 🚩 Red Flag: Below 60%

Net Revenue Retention (NRR)

  • Good Range: 100–120%+

  • 🚩 Red Flag: Below 90%

Logo Retention (Monthly)

  • Good Range: 95–98%

  • 🚩 Red Flag: Below 93%

Radley Finance calculates your unit economics from actual data and benchmarks against industry norms by segment (SMB vs Mid-Market vs Enterprise). When your dataset is too small for confident estimates, it tells you rather than inventing precision.

Explore your unit economics with Radley Finance

This post is the latest installment in a series on fundraising and investment for founders. Previously, we've covered topics like:

Next, we will be looking at what your financial model should look like at each funding stage.