Mathéo Ballasse
Product and B2C distribution expert: he frames the ICP, the go-to-market and the first 60 days for SaaS founders.
Recommendations from our editorial method.
No time to read?
Key takeaways
- AARRR splits your SaaS into 5 stages: acquisition, activation, retention, revenue, referral.
- The trap is optimizing all 5 at once. Unlock the stage that leaks the most first.
- For an early SaaS, the hole is almost always at activation, not acquisition.
You have traffic, a few sign-ups, and still zero steady revenue. You add one traffic source, then another, and nothing moves on the number that counts. That is the classic symptom of a founder optimizing everywhere at once, without knowing where the leak actually is. The AARRR framework exists to fix exactly that: to force you to look at your SaaS stage by stage, and to unlock one thing at a time.

AARRR, the pirate funnel in one sentence
AARRR is the acronym for five stages of a user's journey: Acquisition (how people discover you), Activation (the first moment they touch the value), Retention (they come back), Revenue (they pay) and Referral (they tell others). Investor Dave McClure, founder of 500 Startups, popularized this model in 2007 under the name pirate metrics, because the acronym sounds like a pirate's cry.
The idea behind it is anything but folklore. McClure watched startups drown in vanity metrics (view counts, likes, downloads) without ever knowing whether their product had a future. AARRR answers a single question: at which stage of the journey do your users drop off? Once you know that, you know what to fix.
You don't have five problems to solve. You have one, and it is hiding at a single one of the five stages.
The 5 stages applied to an early SaaS
The generic model talks about a "user journey." Here is what each stage means concretely when you start from zero, with the question to ask yourself and the raw metric to track.
| Stage | The real question | What you measure |
|---|---|---|
| Acquisition | How do people reach your product? | Visitors per channel, cost per visitor |
| Activation | Do they hit their first moment of value? | % of sign-ups doing the key action |
| Retention | Do they come back the next week? | % still active at D+7 and D+30 |
| Revenue | Do they move to the paid plan? | Trial-to-paid conversion rate |
| Referral | Do they tell others about it? | % of newcomers via word of mouth |
The key point: these five stages form a funnel that narrows. Everyone who moves to the next stage is a subset of the previous one. A mediocre number at the bottom can come from a hole much higher up. Here is what an early SaaS journey looks like once you put numbers on it.
These proportions are not made up. The median trial-to-paid conversion rate sits around 8% according to Userpilot's analysis of B2B products, and SaaS retention at 30 days often lands between 20 and 35%. In other words: out of 1000 visitors, a small hundred paying customers is already a decent result, not a failure.
8%
Median trial-to-paid conversion
20-35%
Typical SaaS retention at D+30
20-40%
New customers via referral
That last number is worth pausing on: SaaS companies attribute 20 to 40% of their new customers to word of mouth and referral. The final letter of AARRR is not a decorative bonus, it is a full acquisition channel, free, that most founders neglect because they only think about it once "the rest" is sorted.
The costliest mistake: optimizing everything at once
The natural reflex when you know how to build is to open five worksites at once. You tweak the landing page for acquisition, add onboarding for activation, send win-back emails for retention, test a paywall for revenue, and slap a "refer a friend" on for referral. Result: you move 10% forward everywhere, and therefore nowhere.
Common mistake
A funnel is repaired from the top. If 60% of your sign-ups never reach their first moment of value, every euro spent on acquisition mainly funds... people who are going to leave. You are filling a leaky bucket.
The logic of a funnel is ruthless: the weakest link caps everything that comes after. Doubling your acquisition when your activation is broken means doubling the number of people who drop off at the same spot. AARRR is not a checklist where you tick the five boxes. It is a diagnostic to find the one box to work on right now.
Which stage to start with: find the hole, not the traffic
Here is the counterintuitive part most founders learn too late. When you start out, your problem is almost never acquisition. It is activation. You think you lack visitors when you are losing the ones you already have, for want of a clear, fast first moment of value.

To know which stage to unlock first, run this diagnostic in order. You stop at the first stage that bleeds.
Look at activation first
Then retention
Only then acquisition
Finally, revenue and referral
This discipline changes everything: instead of scattering your energy, you concentrate it on the link that caps the rest. To choose which product action matters most at the activation stage, our guide on the North Star Metric helps you isolate the single indicator that captures perceived value.
Measuring the 5 stages without a machine
You do not need a 300-euro-a-month analytics tool to run AARRR at the start. A spreadsheet and five lines are enough, updated every Friday. One per stage, one raw number, and above all its trend week over week.
My minimal AARRR board
0 / 5The goal is not precision to the decimal, it is to see where it jams and whether it is improving. An activation rate going from 35 to 48% in three weeks tells you your last onboarding change is working. That is infinitely more useful than a perfect dashboard you never look at. To dig into the conversion metric, see our dedicated article on the SaaS conversion rate.
AARRR and acquisition: the link people forget
There is a common misunderstanding. Many read AARRR as a "product" model and file acquisition in a separate box, handled by marketing. Wrong. The first letter of AARRR is precisely acquisition, and it obeys the same logic: one channel done fully beats five lukewarm channels.
Once your activation and retention are healthy, the question becomes: which channel fills the top of the funnel at the best cost? That is exactly the work we break down in our guide on SaaS lead generation, and the natural next step once you have found your first users by hand, as described in finding your first 10 customers. AARRR does not replace your acquisition strategy: it tells you when it becomes profitable.
The pirate framework, at its core, is not one more theory. It is a guardrail against scatter. Five stages, only one to unlock at a time, starting with the one that leaks the most. Do that, and you will stop filling a leaky bucket.
Frequently asked questions
- What does the AARRR framework stand for?
- AARRR is the acronym for five stages of the user journey: Acquisition (how people discover you), Activation (the first moment of value), Retention (people come back), Revenue (they pay) and Referral (they tell others). Dave McClure popularized it in 2007 as pirate metrics to help startups focus on the metrics that actually matter.
- Which AARRR stage should you start with when you launch?
- Not necessarily acquisition. Find the stage that leaks the most in your funnel. Most young SaaS founders think they lack traffic when they are actually losing users at activation, for want of a clear first moment of value. Fix that stage first, otherwise every bit of traffic you add leaks out the same hole.
- Does AARRR work for a B2B SaaS?
- Yes, with adjustments. In B2B the sales cycle is longer, referral takes the form of customer references or partnerships, and retention often weighs more than acquisition in account economics. The five stages stay the same; only the metrics and the timelines change.
Find the stage blocking your SaaS
The diagnostic identifies the funnel stage to unlock first, and the channel that feeds it.