Hard-Bar Activation Metric Selection
Pick the activation metric only 5-15% of users hit, then decompose it into levers
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 95%
Most teams pick an easy activation metric so the number looks good. Isford argues the opposite: choose a metric with a deliberately high bar, one that only 5-15% of users reach, because a rarer state correlates far more tightly with long-term retention. Then decompose that metric into every component that feeds it, and surround it with a small set of supporting metrics so a single number is not making all your decisions.
Origin
Developed by Lauryn Isford with Airtable's activation and analytics teams; the specific metric (week-four multi-user active) was validated by Airtable's analytics team as highly correlated with long-term retention.
Core principles
- 01A low activation rate is a feature, not a bug — it means the bar is meaningfully correlated with retention.
- 02If 40% of users hit your metric, only a fraction of those are still around in month 12, 18, 24. The metric is measuring the wrong thing.
- 03Moving a 5% metric to 6-7% has huge downstream effects on long-term business metrics.
- 04The activation metric is an output metric — the levers live in its components.
- 05One metric never paints the whole picture of a good onboarding; a small portfolio does.
How to run it
- 1
Have analytics find the state that predicts long-term retention
Work with analysts to find the user or team state most highly correlated with long-term retention, rather than picking a milestone by intuition.
Pro tip For a seat-priced, horizontally-adopted collaboration tool, a workspace/team-level metric usually beats a user-level one.
Watch out The right unit of analysis depends on your pricing and adoption motion — a vertically-adopted tool may genuinely need a user-level metric.
- 2
Set the bar deliberately high
Prefer a specific, demanding definition that only 5-15% of users reach over a loose one that most reach. Airtable's was week-four multi-user active: in the fourth week, more than one person on the team is active and contributing to a workflow.
Pro tip A high bar forces multiple things to be true at once — substance built, persistence, collaboration, weekly cadence, real contribution.
Watch out This makes your team's job harder and your headline number smaller. Set expectations with leadership before you adopt it.
- 3
Decompose the metric into every component
Break the metric into its constituent questions: how many sign up as a team vs an individual, how many reach week four, how many invited 2+ people vs had 2+ ever active vs had 2+ active specifically in week four, and what behaviours actually count as 'active'.
Pro tip This decomposition is where you discover which levers can move the aggregate — the output metric itself tells you nothing about what to build.
- 4
Add supporting metrics to escape single-metric tunnel vision
Introduce a small number of complementary metrics — Airtable added an individual retention metric (week-2 / week-4 retained) and 'Build', a sophistication score measuring whether a user reached intermediate usage.
Pro tip These catch the cases where a treatment made users more sophisticated but not more collaborative, or more persistent but not more capable.
Watch out Keep the set small. This is a portfolio, not a dashboard.
- 5
Add guardrail metrics
Pair the North Star with guardrails so a big activation win that quietly damages another part of the business gets caught, and require the team to go deep on any guardrail regression rather than shrug at it.
Pro tip Strong analytics partners raise everyone's literacy on which guardrails matter — PMs, engineers and designers included.
Watch out Onboarding that lifts activation while causing a 10% revenue drop is not obviously a good trade.
In the wild
Airtable's activation team adopted week-four multi-user active as its North Star: in the fourth week after signup, more than one person on a team is active and contributing to a workflow. Before rebuilding onboarding, the team decomposed the metric into its components (team vs individual signup, survival to week four, invites vs ever-active vs week-four-active, and what constitutes contribution) and added retention and 'Build' sophistication metrics alongside it.
→ The decomposition revealed which levers could move activation and made the rebuild targetable; the supporting metrics let the team credit wins on sophistication or retention that the single North Star would have scored as failures.
Common mistakes
Choosing an activation metric because the number looks healthy
A 40% activation rate feels like success but usually means the bar is set at a state that does not predict retention. Only a fraction of those users will be alive at month 12.
Treating the activation metric as something you can build against directly
It is an output metric. Without decomposing it into components, a team has a scoreboard but no levers, and ends up guessing at what to ship.
Making every decision from one number
A single metric cannot express what good onboarding does — it hides treatments that increased sophistication or persistence without moving collaboration, so the team abandons genuinely good work.
Is it for you?
Best for
Growth and activation teams at self-serve/PLG products choosing or resetting their activation North Star
Not ideal for
Very early products with too little data to establish retention correlation, or sales-led businesses where activation is human-driven
From the transcript
“an activation rate that falls in a lower percentage range maybe for most companies 5 to 15 percent is better than one that falls in…”
“for us that was week four multi-user meaning in the fourth week more than one person on that team is active and contributing to a…”
“pick a more specific more precise metric that may be only five percent of users reach but know that those five percent of users will…”
“how many of them make it to week four how many of them have more than two people invited versus more than two people who've…”
“this group of metrics actually really set us free it gave us those constraints where we could actually measure success in a few different ways”
From the episode
Mastering onboarding
Lauryn Isford (Head of Growth at Airtable)