You probably use a dozen SaaS products today without thinking about it: your email, your invoicing tool, your online word processor. What they have in common is not what they do, it is how they are delivered and paid for.
Understanding the model matters directly to a founder, because it produces the most predictable revenue there is, and it also demands the most patience before it produces anything at all.
Definition
SaaS stands for Software as a Service. The software runs on the vendor's servers, users reach it through a browser, and they pay a recurring subscription rather than a one-off licence.
Three traits separate it from conventional software:
- No installation. Users deploy nothing and maintain nothing.
- Continuous updates. Everyone runs the same version, today's.
- Recurring payment. Revenue is measured in MRR (monthly recurring revenue) rather than in sales.
The word gets stretched to cover anything sold by subscription. Monthly access to a template library is not SaaS: there is no software executing anything on the user's behalf.
Why the model appeals
One word: recurrence. In a one-off sales model, every month restarts at zero. In SaaS, last month's customers are still paying this month, and new ones stack on top.
That accumulation changes the nature of the work. You are no longer only trying to sell, you are also trying not to lose, and Retention becomes a growth lever on par with acquisition. A product gaining a hundred customers and losing a hundred every month does not grow, however hard it sells.
The price: the early valley
A €20 monthly subscription earns €20 in month one. A €240 one-off sale earns €240 immediately. It takes a year for the subscription to catch up, and during that year you have to eat.
This is the model's real obstacle, and a badly underestimated one. The growth curves that circulate almost always start at the moment they take off, never during the eighteen flat months before.
Customers stay well past the crossover: the subscription earns more.
Average lifetime is derived from the cancellation rate, of which it is the inverse. Both curves deliberately ignore payment fees, identical either way.
The demo above says it better than any argument: move the price and the subscription length, and watch which month the subscription curve overtakes the one-off sale.
The numbers that steer a SaaS
| Metric | What it tells you |
|---|---|
| MRR (monthly recurring revenue) | This month's recurring revenue, the base of everything |
| Churn rate | The share of customers lost, the hidden brake |
| LTV (customer lifetime value) | What a customer is worth across their whole lifetime |
| CAC (customer acquisition cost) | What they cost to acquire |
| LTV/CAC ratio | Whether acquisition is sustainable or not |
Any one of these read alone tells you nothing useful. Rising MRR with churn rising faster is a product about to stall, and only the combined reading reveals it.
Frequently asked questions
Do I need to code to launch a SaaS?
No, but you need unusual precision about what you are building. No-code tools now ship genuinely functional products, and plenty of profitable Micro-SaaS started that way. The constraint shifts towards running costs as volume grows.
How long before a SaaS is profitable?
There is no honest average, situations vary too much. The better question is different: how many subscribers cover your fixed costs? The Break-even point gives you a number, and that number is often far more reachable than people assume.
Monthly or annual billing?
Both, with a discount on annual. Annual collects twelve months upfront, which solves part of the early cash problem, and it mechanically removes eleven opportunities to cancel. In exchange it makes Churn rate harder to read, since you only measure it once a year.
How is this different from Micro-SaaS?
A difference of scale and ambition, not of nature. Micro-SaaS targets a narrow problem in a small market, with a team of one or two, and accepts that it will never get big. Our Claude Code course shows how to ship a working product without a team, which mainly shortens that early valley.