The Decision Tempo System
Never leave a decision in the room: decide now, pull in whoever is missing, and let deadlines actually bite.
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 7
- Confidence
- 93%
Henrickson's account of how Rippling manufactures decision speed as a system rather than a slogan. The core moves: refuse to schedule a follow-up meeting for a decidable decision, Slack-call the missing decision-maker into the room, hold planning deadlines so hard that missing one means you simply move on, and staff product with people who own domains deeply enough to answer on the spot.
Origin
Jeremy Henrickson describing Rippling's culture under CEO Parker Conrad — explicitly both a personality trait of Conrad's and a deliberate strategic choice he models constantly. Henrickson pairs it with Jeff Bezos-style reversible/irreversible door reasoning ('there are irreversible decisions you can't make that way'), and with the Coinbase debate-then-commit habit he brought from a domain of extreme uncertainty.
Core principles
- 01Most decisions are reversible; treat the irreversible ones as the exception, not the default.
- 02A decision deferred to a scheduled meeting is a week of velocity burned.
- 03If the blocker is a missing person, the fix is to fetch the person, not to postpone.
- 04Deep domain ownership is the substrate of speed: a world-expert PM answers in 30 minutes, not three days.
- 05Deadlines must actually cost something — if a missed date can retroactively force everyone to react, there is no deadline.
- 06In high-uncertainty domains, have the debate loudly, then produce a single company point of view and go full speed at it until you decide to go full speed at a different one.
- 07The culture must permit being wrong out loud, or nobody will commit fast.
How to run it
- 1
Classify the decision as reversible or not
Ask whether this is a genuinely irreversible decision. If it is, slow down and take the process. If it is not — and most are not — it gets made today.
Watch out The failure mode is inflating everything into an irreversible decision to justify delay.
- 2
Decide in the meeting you are already in
Do not schedule a decision meeting for next week, tomorrow, or later today. You are in the meeting; make the decision now.
- 3
Slack-call the missing person
If you cannot decide because a specific person is not present, call them into the room immediately over Slack and be done with the decision today.
Pro tip This only works if the org is flat enough that anyone can pull anyone. Direct exposure to senior leadership is a prerequisite, not a perk.
- 4
Make domain ownership deep enough to answer on the spot
Structure product so a person owns a whole product, not a little feature, and is expected to be the world's foremost expert in it. That expertise is what lets them say 'yes, here's what we should do' immediately, or 'give me 30 minutes to look something up.'
Pro tip See the 'Go and See' framework — going to ground is how a PM earns the right to answer instantly.
Watch out Shallow feature-level ownership guarantees three-day round-trips on every question.
- 5
Bake the tempo into planning deadlines
Run quarterly planning on a timeline that does not leave room for slippage. When someone misses the window, actually move on without them. Do not let a late input retroactively force everyone to re-react.
Pro tip Do it without hostility. It is a norm, not a punishment; people simply have to get used to it.
Watch out It is genuinely shocking to newcomers the first time. If leadership blinks once and reopens a closed decision, the deadline is dead permanently.
- 6
In uncertainty, debate hard then commit hard
Where the future is genuinely unknowable (Coinbase 2017: 'is Ethereum going to be a thing?'), let strong opinions collide fully — then exit the conversation with one clear company point of view. Differing views may persist at the margins; the company goes full speed at the answer until it deliberately chooses a different answer.
Watch out Skipping the debate produces false consensus that unravels under pressure; skipping the commit produces permanent debate.
- 7
Make being wrong cheap to say
Sustain an environment where a leader can say 'I was wrong, here is how I was wrong, let's move on.' Everyone is wrong sometimes; if admitting it is expensive, people will defend bad decisions rather than reverse them fast.
In the wild
Coinbase's usage grew 40x over 2017 while foundational questions ('is Ethereum going to be a thing?') were genuinely unanswered and hotly contested. Security lead Philip Martin reframed each rapid decision in terms of what could be decided quickly while staying secure. The team debated fully, then produced a single company point of view and executed at full speed against it.
→ Coinbase held focus and shipped through the run-up despite systems breaking on the edges most Saturday mornings, without letting unresolved macro debates paralyze execution.
Rippling's quarterly planning timeline leaves little room for decision slippage. When someone misses the date, the company moves on without their input rather than letting them retroactively force everyone to react.
→ Henrickson reports no company he has worked at, from 5 to 5,000 people, has ever operated at Rippling's tempo — and the principle now reinforces itself by its own gravity rather than by enforcement.
Common mistakes
Treating decision speed as a values poster
Speed here is produced by concrete mechanisms — Slack-calling people mid-meeting, unmovable planning dates, whole-product ownership. Declaring 'we move fast' without those mechanisms changes nothing.
Reopening a decision for a latecomer
Allowing someone who missed the deadline to make everyone react retroactively destroys the deadline permanently. The move-on has to be real and it has to be non-hostile.
Making irreversible decisions at reversible speed
Henrickson is explicit that some decisions cannot be made this way. The system depends on correctly classifying the small set that deserve slowness.
Is it for you?
Best for
Founders and execs at fast-growing companies whose velocity is being eaten by decision latency rather than execution capacity, and who have (or can build) flat access to decision-makers.
Not ideal for
Highly regulated one-way-door decisions, safety-critical systems, or organizations where deep domain ownership does not yet exist — speed without expertise just produces fast bad calls.
From the transcript
“let's like slack call in the person that we need in order to make that decision and we'll be done with the decision today and…”
“feature you own like your product and you're expected to be the world's foremost expert in it and if you are what that means is…”
“it's shocking to people when we actually move on right that haven't been here yet it's like no no that date passed”
“you go full speed toward this answer until you decide to go full speed toward a different answer”
From the episode
Moving fast and navigating uncertainty
Jeremy Henrickson (Rippling, Coinbase)