Example data. Fieldnote is a fictional company used to fill all 61 boards, so the examples tell one consistent story.
Post-PMF Fieldnote, the quarterly AARRR review with Karim, the data analyst, Nadia and Élise, looking for the next growth lever.
What the team took away
The funnel leaks at activation, not acquisition: the team cuts paid search and puts the quarter on time-to-first-10-interventions and month-1 invoicing connection.
How to run it
Preparation
Product, data, and growth. One person must be able to pull numbers live; if not, the session becomes a debate about definitions.
Agree the definition of each funnel step one week before: what counts as activated, what counts as retained at day 30. Write them on the board.
Pre-fill the five steps with last month's numbers and the conversion rate between each step. Bring the same numbers for the previous quarter.
Agenda
15 min
Read the funnel Walk the five steps with the numbers. No interpretation yet, just confirm everyone reads the same figures and definitions.
25 min
Find the leak Compare each conversion rate to the previous quarter and to what you would expect. Circle the single step where the drop is biggest relative to its potential.
30 min
Segment the leak Data cuts the leaking step by segment: acquisition channel, plan, device, cohort. Look for one segment that behaves differently from the rest.
30 min
Hypotheses Write hypotheses on why the segment leaks. Each one must name a cause you could act on within a month. Vote with three dots each.
20 min
Commit Pick the top hypothesis. Define the experiment, the metric it should move, and the number that would count as success. Owner, review date in 4 weeks.
Pitfalls
Do not work on two steps at once. A funnel review that fixes retention and activation in the same quarter fixes neither.
Do not accept a drop without a segment cut. An average leak hides one segment that leaks a lot and others that do not.
Do not leave with a hypothesis and no number. 'Improve activation' is not a commitment; 'activation from 32 to 38 percent for organic signups' is.
Afterwards
The chosen hypothesis goes onto the experiment board and the number becomes an input metric on the North Star tree. Rerun the review monthly; change the target step only when the current one has moved.