The Gut-Check Protocol
Intuition is pattern matching and bias in the same organ — check it with System 2 and diverse guts.
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 4
- Confidence
- 90%
Antin's answer to the trust-your-gut vs. be-data-driven false choice. Intuition is real and necessary — it's the pattern-matching output of experience — but it is also exactly where every cognitive bias and blind spot lives. The protocol: keep the gut, then check it two ways. First with System 2 (slow, analytic, methodical — but slow meaning hours, not months). Second with a genuine wisdom-of-the-crowd, which only works when the people in the room bring genuinely diverse sources of information and judgment.
Origin
Judd Antin, drawing explicitly on Kahneman's Thinking, Fast and Slow (System 1 / System 2), the wisdom-of-crowds literature, and Duncan Watts's 'Everything Is Obvious: Once You Know the Answer' on hindsight bias and the narrative fallacy. Taught by Antin in a leadership class at UC Berkeley.
Core principles
- 01The gut is where heuristics live — and where all the cognitive biases live too.
- 02System 2 is not sexy and not months long: it's a deliberate, slow-ish analytic pass, not a split-second call.
- 03The wisdom of the crowd only works when the crowd holds diverse information; identical inputs produce no wisdom at scale.
- 04Hindsight bias ('we knew that already') is a form of self-gaslighting via the narrative fallacy.
- 05A great researcher's job is to keep expanding your horizons so your intuition keeps improving.
How to run it
- 1
Name the gut call explicitly
Surface the intuitive judgment as a claim before testing it. Intuition earned from pattern matching is legitimate input — but it must be stated to be checkable.
Watch out There are 151 catalogued cognitive biases; you will not out-think them by trying harder inside System 1.
- 2
Run a System 2 pass
Engage the slow, methodical, analytic process as a deliberate check on the gut. Slow relative to a split-second decision — not months of analysis.
Pro tip Time-box it. The point is deliberateness, not duration.
- 3
Assemble diverse guts, not more guts
Get people in the room who bring different sources of information and judgment — evidence and intuition both — and run a direct, kind conversation where disagreement is expected. Same-information crowds add no wisdom regardless of headcount.
Pro tip Researchers are unusually good at leading these sessions — surfacing and reconciling conflicting evidence is the job description.
Watch out A room of people who all read the same doc is not a crowd.
- 4
Look for the collision
The valuable moment is when your intuition and the evidence collide — affirming or falsifying the decision. That collision, not consensus, is where good things happen.
Watch out Guard against the 'we knew that already' reflex when evidence merely affirms — that's hindsight bias covering for a decision you'd have made either way.
In the wild
Antin's illustration: ask anyone how they got to where they are and they will produce a clean causal series of events. That story is probably a lie — we selectively remember and construct narratives that make the past feel obvious and inevitable.
→ The same machinery makes teams say 'that was obvious, we knew that already' about research findings they in fact did not know.
At Facebook, an assumption that hidden posts signalled bad content was checked by putting decision-makers behind the mirror for a single interview with a heavy hider, who turned out to be chasing inbox zero on an infinite feed.
→ An n-of-1 collided with the room's intuition and killed a naive signal — no statistical significance required, because the deciders were present.
Common mistakes
Polling a homogeneous room and calling it wisdom
Wisdom of the crowd depends on diversity of information and judgment. If everyone shares the same inputs, adding people just amplifies the shared bias.
Treating System 2 as a months-long analysis project
Antin is explicit that slow means deliberate, not lengthy. Turning the gut-check into a research programme reproduces the 'research slows us down' complaint.
Is it for you?
Best for
Founders, PMs, and executives who make high-stakes product calls on instinct and want a cheap bias check that doesn't stall the decision
Not ideal for
Genuinely reversible, low-stakes decisions where the cost of the check exceeds the cost of being wrong
From the transcript
“so the secret is not that sexy it's system too right so you engage that slow methodical process in which you do analytic thinking as…”
“the wisdom of the crowd works when the people involved with the decision are bringing diverse sources of information and judgment to the table”
“so if you want to check your gut get a bunch of different guts together”
“they need to be open to the fact that your gut is limited and biased and and narrow and wrong sometimes”
From the episode
The UX research reckoning is here
Judd Antin (Airbnb, Meta)