The Galvanizing Activation Metric
The activation number's job is organizational momentum, not statistical precision.
- Difficulty
- Easy
- Time to result
- ~weeks to results
- Steps
- 4
- Confidence
- 90%
Facebook's famous '10 friends in 14 days' was not brilliant because the number was correct. Adriel Frederick's reading is that it was brilliant because it was designed to be understood, remembered, and chased. A generic goal ('get people as many friends as possible as fast as possible') produces nodding; a discrete number, a discrete time window, and a graph everyone can watch go up produce coordinated action. The metric's function is galvanizing, and its precision is nearly irrelevant.
Origin
The 10-friends-in-14-days target predates Frederick at Facebook (he says Zuck ended the internal debate by simply declaring the numbers); Frederick's contribution is the framework explaining why it worked and how to design one, including its predictable side effects.
Core principles
- 01The value of an activation metric is that it's memorable and communicable, not that it's statistically optimal.
- 02A discrete number plus a discrete time window plus a public graph produces galvanizing organizational effect.
- 03Generic goals get agreement but not action; concrete goals get chased.
- 04Ending the academic debate is a leadership act, not an analytics act.
- 05Directionally correct beats precisely correct.
How to run it
- 1
State the underlying growth truth generically
Write the real objective in plain terms — e.g. 'get people as many friends as possible as fast as possible.' Notice that stated this way, everyone agrees and nobody acts.
- 2
Convert it into a discrete number and a discrete time window
Pick a concrete count and a concrete deadline (10 friends, 14 days). The pairing creates a chaseable goal instead of a direction.
Pro tip The number should be defensible and roughly correlated with retention — but stop optimizing it well before the debate feels settled.
Watch out Endless debate about the exact number is the failure mode. At Facebook, Zuck ended it by declaring '10, 14 days, go.'
- 3
Put it on a graph everyone can watch
Make it visible and public inside the company so every team can see the line and know they are collectively pushing it up. Visibility is what converts a target into momentum.
- 4
Accept the distortions the metric creates
Teams will optimize against the number, sometimes literally — Frederick recalls a retention-vs-friend-count graph with a kink at 10 because someone had hard-coded 10 as a limit somewhere, so retention dropped at 11 or 12 before recovering. Judge the distortion against the counterfactual: without the organizational momentum, the whole curve would have been lower.
Pro tip When you find a Goodhart artifact, ask whether the level of the entire curve is higher because of the metric. Usually it is.
Watch out This tolerance applies to benign artifacts, not to metrics that drive genuinely harmful behavior.
In the wild
Facebook debated at length what the friend count and time frame should be. Zuck ended it: '10, 14 days, go.' The number wasn't magic and nobody claims it was the true retention inflection point. But it was memorable, communicable, and gave every team one graph to push up.
→ The whole organization got past the academic debate and started chasing the goal. Frederick notes the metric is still being talked about a decade later — proof of its design for memorability.
Looking at a graph of retention versus number of friends, Frederick saw retention actually drop at 11 or 12 friends versus 10 — because somewhere in the code someone had hard-coded 10 friends as a limit while trying to improve retention. The curve recovered above 12.
→ Frederick judged the artifact acceptable: without the organizational momentum the metric created, the entire retention curve would have sat lower. He took the kink.
Common mistakes
Chasing the perfect number
Teams burn months debating the precise count and window as if a correlation coefficient will settle it. The precision buys nothing; the delay costs everything.
Setting a generic directional goal
'Get users more friends faster' earns agreement and produces no coordinated action. Without a discrete number, a discrete window, and a graph, there is nothing to chase.
Panicking at the first Goodhart artifact
Metrics get gamed and produce odd kinks. If the entire curve is higher because the organization was galvanized, the artifact is a fair price.
Is it for you?
Best for
Growth leaders and founders who need to align a whole organization on one activation goal and are stuck in analytics debate.
Not ideal for
Domains where optimizing a proxy metric carries real user harm (safety, health, financial risk), where precision and guardrails must come first.
From the transcript
“it was the designing it to be understood and communicates it”
“discrete number and a discreet time and there is a concrete goal to cheese and there's a number and a graph that everybody can look…”
“at some point Zuck just said 10 14 days go”
“if we didn't get that like organizational momentum that graph would have just been lower so I could take the Kink where it drops it's…”
From the episode
Humanizing product development
Adriel Frederick (Reddit, Lyft, Facebook)