Sign inGet started
Skills / Startup-growth

AARRR Framework

Most startups are focused on the wrong metrics. They obsess over Acquisition when their real problem is Activation — users show up, look around, and leave before they ever experience the product's core value.

The AARRR Framework ('Pirate Metrics') is Dave McClure's five-stage funnel model for startup growth: Acquisition, Activation, Retention, Referral, and Revenue. It gives founders and growth teams a repeatable diagnostic process — identify your leakiest stage, then run targeted experiments to fix it before moving on. Unlike generic growth advice, AARRR forces a sequential funnel audit that tells you exactly where to focus and why.

By Dave McClure · Free
Specimen 01 · Live diagnosisAARRR Framework
Input

“B2B SaaS. We sell project management software to small agencies. We've been running ads and our traffic is decent but revenue growth has stalled.”

Diagnosis
Let's run a Pirate Metrics audit on your startup.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

Stop spending on acquisition — find and fix the stage where your startup actually leaks

AARRR maps startup growth into five sequential, measurable stages. Acquisition covers how users find you (channels, CAC). Activation measures whether users have a meaningful first experience — the 'aha moment' — within their first session. Retention tracks whether users return (DAU/MAU, cohort curves, churn). Referral measures word-of-mouth and viral mechanics (NPS, viral coefficient). Revenue captures monetization events (ARPU, LTV, conversion to paid). The core diagnostic exercise is the 'leaky bucket': calculate inter-stage conversion rates, identify the steepest drop-off, and fix that stage before investing more in earlier stages. McClure's central insight is that most startups over-invest in Acquisition while Activation rates below 10% silently destroy all the money they're spending.

The problem

Most startups die not because they can't get users, but because they can't keep them — or never activated them in the first place. The AARRR Framework addresses the chronic startup mistake of spending on Acquisition channels while Activation rates sit below 10% and month-one retention is near zero. Without a sequential funnel audit, teams optimize the wrong stage and burn budget with nothing to show for it.

The solution

Walk away with a clear diagnosis of your leakiest funnel stage, a tailored KPI set for your business model, and a prioritized experiment backlog so you know exactly what to test next — not just 'grow faster.'

You bring
  • Your current metrics or best estimates for each funnel stage (even rough numbers work)
  • A description of your business model (SaaS, marketplace, consumer app, e-commerce)
  • A sense of which stage feels most painful right now
  • Willingness to be honest about stages where you're not measuring anything yet
You get
  • A funnel conversion rate table showing where users are dropping off most severely
  • A clear diagnosis of your primary leaky stage with the reasoning behind it
  • A KPI definition set tailored to your specific business model for all five stages
  • A prioritized list of growth experiments targeting your weakest stage
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“You're spending $8,000/month on Google Ads, getting 300 signups, and converting 15 to paid. Your team is debating whether to increase ad spend, try new channels, or rebuild the pricing page. Nobody has looked at what percentage of signups actually complete onboarding, and you have no idea whether users who churn in month two are different from those who stay.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With AARRR Framework

“Your funnel audit reveals a 13% Activation rate — the real leak. You pause the ad budget increase, run three onboarding experiments over six weeks, and push Activation to 28%. The same 300 signups now produce 34 paid conversions instead of 15. You have a defined KPI for each AARRR stage, a cohort retention chart showing month-by-month behavior, and a growth backlog ranked by stage priority so every sprint has a clear focus.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
AARRR FrameworkClaude
AARRR Framework
Let's run a Pirate Metrics audit on your startup. I'll take you through all five AARRR stages — Acquisition, Activation, Retention, Referral, Revenue — and we'll find your leakiest bucket. To calibrate the right metrics for your business: are you B2B SaaS, consumer app, marketplace, e-commerce, or something else?
B2B SaaS. We sell project management software to small agencies. We've been running ads and our traffic is decent but revenue growth has stalled.
Got it. Before we talk about the ads, let's check the downstream stages — that's almost always where the real problem hides. Starting with Activation: what's your definition of an 'activated' user? For B2B project management, that's typically something like 'created their first project and invited at least one teammate' within the first session or first week. Do you track anything like that, and if so, what percentage of signups hit that milestone?
Honestly, we've never defined it that way. We just track signups. We get maybe 300 signups a month from the ads but only about 15-20 end up converting to paid.
That's a 5-7% Acquisition-to-Revenue conversion — but the real question is where the 93% goes. With B2B SaaS, the most common culprit is Activation: users sign up, get confused or overwhelmed in the first session, and never return. A 300-signup month with 15 paid conversions suggests your Activation rate might be very low. Let me ask: do you have any data on what users do in their first session? Specifically — what percentage create a project, invite a teammate, or complete any 'setup' action?
We do have Mixpanel but we've never looked at this. I just pulled it up — looks like about 40 out of 300 signups actually create a project in their first week. That's like 13%.
There's your leaky bucket. A 13% Activation rate means 87% of your ad spend is wasted the moment users hit your product — before Retention, Referral, or Revenue even have a chance to work. The diagnosis: your primary problem is Activation, not Acquisition. Here's the implication: if you doubled your ad spend tomorrow, you'd get 600 signups and maybe 30 paid conversions, but you'd still be throwing away 87 cents of every dollar. The leverage move is fixing Activation first. For B2B project management, the highest-impact experiments are usually: (1) an interactive onboarding checklist that guides users to 'create project + invite teammate' in session one, (2) an automated Day 1 email with a single CTA to complete setup, and (3) a welcome call trigger for signups from high-intent channels. Want me to build out a prioritized experiment backlog for your Activation stage, or should we first finish the full funnel audit to confirm there's no worse leak downstream?
Let's finish the audit first. That makes sense.
Reply to AARRR Framework
What the skill is doing
01AARRR Funnel Stage Audit
02Leaky Bucket Diagnosis
03Stage-Specific KPI Builder
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
AFAARRR Frameworktyping…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
B2B SaaS. We sell project management software to small agencies. We've been running ads and our traffic is decent but revenue growth has stalled.9:14✓✓
Let's run a Pirate Metrics audit on your startup.9:14
I'll take you through all five AARRR stages — Acquisition, Activation, Retention, Referral, Revenue — and we'll find your leakiest…9:15
To calibrate the right metrics for your business: are you B2B SaaS, consumer app, marketplace, e-commerce, or something else?9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

AARRR Funnel Stage Audit

A structured interview that walks through each of the five stages sequentially, surfacing your current metrics, the tools you're using to measure them, and the gaps where you're not measuring at all. The audit surfaces the full funnel picture before any diagnosis is attempted.

Based on McClure's five-stage sequential model from 'Startup Metrics for Pirates' — each stage must be assessed individually because problems in later stages (Activation, Retention) are invisible when teams only report top-of-funnel Acquisition numbers.
CapabilityC-02

Leaky Bucket Diagnosis

Calculates or estimates inter-stage conversion rates and identifies the single stage with the steepest drop-off — the 'leaky bucket.' Explains why fixing this stage delivers the highest leverage before investing in earlier stages.

The leaky bucket metaphor is central to McClure's framework: pouring more water (Acquisition spend) into a leaking bucket wastes resources. The framework's primary diagnostic heuristic is to always fix the worst leak first.
CapabilityC-03

Stage-Specific KPI Builder

For a given business model, recommends the right metrics for each of the five stages — not generic metrics, but the specific ones that matter for SaaS recurring revenue, marketplace liquidity, consumer mobile engagement, or e-commerce repeat purchase.

McClure's framework explicitly requires business-model-specific metric selection. Activation for a B2B SaaS product (first workflow completed) differs entirely from a consumer app (first share or saved item). The KPI Builder applies this contextual branching.
CapabilityC-04

Growth Experiment Backlog Generator

Once the leaky stage is diagnosed, generates a prioritized list of growth experiments targeting that specific stage — including A/B test ideas, onboarding changes, retention triggers, referral mechanics, or pricing tests depending on the stage.

McClure's methodology is explicitly iterative: diagnose the weakest stage, run experiments to improve it, then reassess the full funnel. The experiment backlog operationalizes this loop by providing stage-appropriate experiment templates.
CapabilityC-05

Business Model Metrics Customization

Adapts the five AARRR stage definitions and their metric categories to the user's specific business model, preventing the common mistake of applying SaaS metrics to a marketplace or consumer mobile metrics to a B2B product.

A known challenge with AARRR is that stage definitions blur across business models. McClure's framework requires practitioners to define what 'Activation' means for their specific product — this capability ensures that customization happens explicitly before any diagnosis.
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

AARRR Funnel Conversion Report

A five-stage funnel table showing your current metrics, inter-stage conversion rates, and a highlighted diagnosis of the leakiest stage with the supporting rationale. The single document that anchors all growth prioritization decisions.

OutputD-02

Stage-Specific KPI Dashboard Spec

A structured specification of 2-3 primary KPIs per AARRR stage, customized for your business model, with definitions, measurement methods, and benchmark ranges where applicable. Ready to hand to a data analyst or build in your analytics tool.

OutputD-03

Prioritized Growth Experiment Backlog

A ranked list of growth experiments targeting your diagnosed leaky stage, with experiment type, hypothesis, success metric, and implementation complexity for each. Designed to feed directly into your sprint planning or growth meeting.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Dave McClure

Dave McClure is the creator of Pirate Metrics and co-founder of 500 Startups, one of the world's most active early-stage venture funds with investments in 2,000+ companies across 75 countries. He introduced the AARRR framework in a widely-circulated 2007 SlideShare presentation that became foundational curriculum for YC, Techstars, and accelerator programs globally. His growth methodology has shaped how an entire generation of startup operators thinks about metrics.

Status · Inspired by Dave McClure’s work — not yet claimed. Are you Dave McClure?
Primary sourceS-01

Startup Metrics for Pirates (2007 SlideShare presentation)

by Dave McClure

Co-founder, 500 Startups (2,000+ portfolio companies, 75 countries); creator of the AARRR Pirate Metrics framework; original 2007 deck has been viewed millions of times and remains a standard reference in startup accelerator curricula.

Read the original ↗
Citationslideshare.net
In the build queue

Be first to run it.

AARRR Framework is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail
At launchI want to run an AARRR audit on my startup. We're getting decent top-of-funnel traffic but revenue growth is flat and I'm not sure where we're leaking. Can you walk me through the Pirate Metrics framework stage by stage?