How to fill in the AARRR Pirate Metrics
Acquisition to referral: find the one funnel stage that is quietly leaking. This guide walks every block in the recommended order — what belongs there, the questions that unlock it, and patterns from real canvases.
AARRR breaks a product's growth into five funnel stages — Acquisition, Activation, Retention, Revenue, Referral — and asks you to attach one honest metric to each. Dave McClure proposed it as an antidote to vanity metrics: page views and signups feel good, but only stage-by-stage conversion tells you where growth actually dies.
The discipline is one metric per stage, each with a current value. A funnel with numbers immediately shows its weakest stage, and the weakest stage is the only one worth working on — improving acquisition while activation loses 80% of arrivals is paying to fill a leaking bucket.
Created by Dave McClure (500 Startups) in his 2007 talk 'Startup Metrics for Pirates' — the name comes from saying AARRR out loud. It remains the default funnel model for SaaS and consumer product teams.
Acquisition
How do people find us, and what does each channel cost?
Every channel that brings visitors: search, content, paid, social, PR, partnerships. For each, note volume, cost per visitor, and — critically — quality, measured by how well that channel's visitors convert downstream. The best acquisition channel is rarely the biggest one; it is the one whose users still exist at retention.
Ask yourself
- Which three channels bring most visitors, and at what cost each?
- Which channel's users survive to activation and retention best?
- What is our cost per acquired user, by channel?
- Which channel are we funding out of habit?
Patterns that work
- Judge channels by downstream quality, not arrival volume
- Cost per acquired user, not cost per click
- One number to own: visitor → signup conversion
Activation
Do new users reach the moment of first value?
Activation is the first experience of real value — not the signup, the moment the product does what they came for. Define that moment precisely for your product (first report generated, first message sent, first search that worked), then measure what fraction of signups reach it and how long they take. This is the most commonly broken stage and the cheapest to fix.
Ask yourself
- What exactly is our 'aha' moment, defined as a user action?
- What percentage of signups reach it — and within how long?
- Where in onboarding do most people stall?
- What could we remove from the path to first value?
Patterns that work
- Define activation as an action, never as 'completed onboarding'
- Measure time-to-value in minutes or days, then halve it
- Fixing activation multiplies every pound already spent on acquisition
Retention
Do they come back without being pushed?
The proportion of users still active after a week, a month, a quarter — measured by cohort, because a blended average hides everything. Retention is the most honest stage in the funnel: it cannot be bought, and it compounds. A retention curve that flattens means a real product; one that slides to zero means the other four stages are decorating a leak.
Ask yourself
- What does week-4 retention look like, by signup cohort?
- Does the retention curve flatten, and at what level?
- What do retained users do that churned users did not?
- Which moment predicts long-term retention?
Patterns that work
- Cohorts, not averages — blended numbers hide the trend
- A flattening curve is the strongest product-market-fit signal there is
- Find the behaviour retained users share, then drive new users to it
Revenue
Where does money actually enter, and what is a user worth?
Conversion to paying, average revenue per user, and lifetime value against acquisition cost. The stage-level question is whether the unit economics close: LTV comfortably above CAC, and payback inside a survivable window. If they do not, growth at the top of the funnel makes the business worse, faster.
Ask yourself
- What fraction of active users pay, and what triggers the upgrade?
- What are LTV and CAC, honestly calculated?
- How many months until a customer pays back their acquisition cost?
- What would users happily pay more for?
Patterns that work
- Identify the upgrade trigger — the moment free stops being enough
- LTV built from real retention, not hoped-for retention
- Expansion revenue is the cheapest revenue you will ever find
Referral
Do users bring other users?
How many new users each existing user generates — through invitations, sharing, word of mouth, or the product being visible in use. Referral is the only stage that lowers acquisition cost as it grows. Be honest about whether your product has natural share moments; a referral programme bolted onto a product nobody mentions produces coupons, not growth.
Ask yourself
- How do new users say they heard about us, actually?
- Where is the natural share moment in the product?
- What is our viral coefficient — invites sent × conversion?
- Would users recommend us unprompted, and have they?
Patterns that work
- Find the moment worth sharing before building the incentive
- Viral coefficient = invites per user × conversion per invite
- Referral quality beats referral volume — track their retention too
Ready to fill yours in?
The editor carries this whole guide with it — every block has these prompts and starter notes built in. Free, no signup, autosaves in your browser.
Open the AARRR editor