From the field
AARRR
A measurement structure that splits the customer journey into acquisition, activation, retention, referral, and revenue, and assigns exactly one metric to each stage so no stage can borrow success from another.
The default agency report measures what platforms make easy: impressions, clicks, CTR. All of it lives in the first stage of the journey, which is why a report can look excellent while the business shrinks. AARRR forces one number per stage, so the leak has nowhere to hide.
Use it when
- Client reports are full of clicks and impressions but no revenue line
- You need to find which stage of the funnel is actually leaking
- Designing a monthly report that survives a sceptical owner
Do not use it when
- True one-off purchases with no retention or referral loop; trim the stages rather than invent fake ones
- Brand-new accounts with too little data to fill five stages honestly
The steps
- Define the five stages for this business. Acquisition, activation, retention, referral, revenue, in the client's own words.
- Assign exactly one metric per stage. More than one and the report is negotiable again.
- Fill in a month of real data. Gaps are findings: an unmeasured stage is an unmanaged stage.
- Fix the worst stage first. The leak furthest down the funnel wastes everything spent above it.
AARRR
- Acquisition
- Activation
- Retention
- Referral
- Revenue
Worked example
A dental clinic spending €2,000/month on Google Ads, illustrative numbers:
| Stage | Metric | Value |
|---|---|---|
| Acquisition | New patient enquiries | 46/month |
| Activation | Enquiries that attend a first appointment | 61% |
| Retention | Patients returning within 12 months | 38% |
| Referral | New patients naming an existing one | 4/month |
| Revenue | Revenue per new patient, year one | €410 |
The 61% activation rate was the finding: 18 paid-for enquiries a month never became patients. Fixing phone handling was worth more than any change inside the ad account.
Why it works
One metric per stage removes the reporting ambiguity that lets everyone declare success. It also relocates problems correctly: half of what looks like a marketing problem is an operations problem, and this is the cheapest way to prove which. Where AARRR assigns one metric per stage of a journey the customer has already entered, See-Think-Do-Care does the same for the intent stages before it, and the two report cleanly side by side.
Who this is for
The same method, read three ways.
01
Running it
One metric per stage, and no stage allowed to borrow another's success. It is the structure that stops a good acquisition number hiding a retention problem.
02
Teaching it
A compact way to show why a single headline metric misleads. Assigning exactly one measure per stage makes the trade-offs between stages visible in a way a dashboard rarely does.
03
Measuring your own funnel
Five numbers that tell you which part of the business is actually broken. Usually it is not the part you are spending money on.