SaaS Product Metrics in 2026: What to Track and When Your Build Pays Back

A SaaS founder doesn’t need twenty metrics. Before launch, track one: how many new users reach the first useful result. After launch, track monthly recurring revenue, churn and four unit numbers: profit a month per customer, lifetime value, the cost to win a customer and the month the build is paid back. Survey scores like NPS help you find problems once people have used the product for a while. In 2026, put payment and app store fees into your margin from day one. Leave them out and every number after that looks better than it is.

Which metrics matter at which stage

Stage	Track	Leave for later
Before launch: prototype or beta	Activation rate, time to the first useful result	ARR, NPS
First paying customers	Trial-to-paid conversion, MRR, customer churn	Expansion revenue
Growing	Revenue churn, lifetime value, cost to win a customer, payback month	Sign-up counts on their own
Raising money	ARR and its growth, churn by group of customers who joined in the same month	Anything you can’t show from your billing data

The formulas we use

A few formulas in the earlier version of this post were off. These are the ones we use.

• Activation rate = new customers who reached the first useful result ÷ all new customers × 100.
• Trial-to-paid conversion = trials that became paying customers ÷ trials started × 100.
• Customer churn (monthly) = customers lost during the month ÷ customers at the start of the month × 100. Count customers lost, not the net change, or new sign-ups will hide the losses.
• Revenue churn (monthly) = MRR lost to cancellations and downgrades ÷ MRR at the start of the month × 100.
• MRR = the sum of monthly subscription revenue from paying customers. ARR = MRR × 12.
• Cost to win a customer (CAC) = sales and marketing spend ÷ new paying customers won in the same period.
• NPS = % of promoters (9-10) minus % of detractors (0-6).
• CSAT = positive answers ÷ all answers × 100.

Unit economics: when the build pays back

These are the equations our project estimate shows. They are simple enough to check by hand.

• Profit a month = price a month × margin.
• Lifetime value = profit a month × the months a customer keeps paying.
• Left per customer = lifetime value − the cost to win a customer.
• Customers who pay for the build = build cost ÷ what is left per customer.
• Payback month: start at minus the build cost. Each month, add the profit from everyone still paying and subtract what you spent to win new customers. The payback month is the first month the running total reaches zero.

For the ratio of lifetime value to the cost of winning a customer, 3× or more is healthy. Above 10× is rare, so test the cost to win a customer first.

Here is an example run of our estimate’s math with sample inputs for a B2B SaaS web app with teams, roles and subscription billing. Every input is an assumption. Only the arithmetic is ours.

Starting number (assumption)	Value	Result	Value
Price a month	$99	Profit a month	$79
Margin	80%	Lifetime value	$1,896
Months a customer pays	24	Left per customer	$1,296
Cost to win a customer	$600	Lifetime value ÷ cost to win	3.2×
New customers a month	15	Customers who pay for the build	8
Build cost (middle of a $6,600-13,600 range, 5 weeks)	$10,100	Payback month	16

Why 8 customers but month 16? Because customers arrive 15 a month and you pay $9,000 a month to win them before they have paid much back. In this run, month 8 is the first month that earns more than it spends, and the running total turns positive in month 16. Change the price or the cost to win a customer and both months move.

The fees that cut your margin in 2026

Margin is what is left after the costs of serving a customer, including payment fees. These are the current listed fees, from each company’s own page:

• Stripe lists 2.9% plus a fixed fee for a domestic card payment, more for international cards, and 0.7% of billing volume for its pay-as-you-go Billing product.
• Paddle lists 5% + 50¢ per checkout transaction. As the merchant of record, it also collects and files sales tax.
• Apple charges 15% under its Small Business Program to developers new to the App Store or with up to $1 million in proceeds the year before. On Apple’s alternative terms in the EU, the program rate and the rate on subscriptions after their first year drop to 10%.
• Google Play charges auto-renewing subscriptions under the first $1 million of yearly earnings 10% plus a 5% billing fee in Australia, the EEA, Japan, the UK and the US, and 15% in other markets.

A subscription sold through an app store can lose 15% before you pay for servers or support. Model it.

Market size: count customers you can reach

For market size, the estimate suggests customer groups and gives each a TAM, SAM and SOM with a note on what the number covers. Treat them as hypotheses to test, not verified demand. Then check the SOM against a count you build yourself:

Step	Input	Result
Reachable prospects in one group	2,000 independent clinics in one country	2,000
Share you can win in years 1-3	5%	100 customers
Price a month	$99
Revenue a year	100 × $99 × 12	$118,800

If the SOM in your estimate is far above a number built like this, trust the bottom-up number. Our guide to SaaS financial models walks through the rest of the model.

What we have seen in real products

• GhostMinutes reached $1,500 MRR. After launch, the founder tested different ways to charge. The live product now has four plans, Free, Lite, Pro and Business, paid monthly or yearly, plus minute packs paid once.
• IPOPBOOKS launched with a 10-day free trial with no card, or payment from day one, then one of three monthly plans.

What a SaaS build costs

A first release typically costs $5,000 to $15,000 with us and takes 2 to 4 weeks; see SaaS development. Across 95 agencies, the median published MVP starting price is $13,234, and the middle half of starting prices runs from $6,617 to $20,000 (Rocketech MVP Price Index 2026, H2 edition).

How we would estimate yours

1. Our project estimate turns your brief into user stories. You confirm what is in scope, and every hour in the budget traces back to a story you kept.
2. We check the hours the model suggests against what our team actually books on finished projects, then multiply by our published rate of $39-55 an hour.
3. For the customer groups you keep, the unit economics show how many customers pay for the build and in which month it is paid back. You can change any starting number and the math updates.

It takes about four minutes to see what your SaaS costs to build and when it pays back. It is a planning range, not a fixed quote.

What we would do

1. Write down the activation event before launch: the first action that shows a customer got value.
2. Put payment or app store fees into your margin.
3. Measure the cost to win a customer for each channel, every month.
4. Track churn by the month customers joined, so a good month of sign-ups doesn’t hide a leak.
5. Recompute the payback month every month. If lifetime value minus the cost to win a customer is zero or less, fix the price or the cost to win a customer before you build more.
6. Ask for NPS after customers have used the product for a while, not on day one.

FAQ

Which SaaS metrics matter most for an early-stage startup?

Before launch, the activation rate. After the first paying customers, MRR, customer churn and trial-to-paid conversion. Once you spend on growth, add lifetime value, the cost to win a customer and the payback month.

How do you calculate customer lifetime value?

Multiply the profit a month per customer (price × margin) by the months a customer keeps paying. At $99 a month and an 80% margin, profit is $79 a month; over 24 months, lifetime value is $1,896.

What is a good LTV to CAC ratio?

3× or more is healthy. Above 10× is rare, so check the cost to win a customer before you trust it.

How many customers do I need to pay back the build?

Divide the build cost by lifetime value minus the cost to win a customer. In our example run with sample inputs, a $10,100 build needs 8 customers, and the payback month is 16 because customers arrive over time.