The Right Kind of Stubborn Test
A short set of honest questions for deciding whether to persist or to change your mind.
- Difficulty
- Moderate
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 83%
Sean Puri names this as the hardest question a founder faces: am I being the right kind of stubborn? The two canonical stories cancel out. Pinterest had no traction for over a year until the founder walked into an Apple store and set Pinterest as the default homepage on the machines, which reads as pure determination beating the data. The opposing advice is to be data-driven and feedback-oriented, listen to the market, and pivot if nobody wants what you built. Sean's position is that no generic rule resolves this, so you replace the rule with a set of questions you answer honestly: has what I originally believed about this actually changed, do I have genuinely new information that should update my thinking, and am I holding this position only so I do not have to feel wrong. The same test governs investments — it is what led him to buy more Facebook while the stock was getting crushed, and to reassess his Bitcoin-and-inflation thesis when inflation spiked and people moved to dollars anyway.
Origin
Sean articulated it while Bloom pressed him on how to avoid drifting from being a resilient 'local' into a stubborn one who refuses to update.
Core principles
- 01There is no generic answer — it is context and circumstance dependent
- 02Test the thesis, not the price or the traction number
- 03New information matters; the same information feeling worse does not
- 04Ego-protection is the most common hidden reason for holding
- 05Being wrong is a state you must be willing to accept out loud
How to run it
- 1
State the original thesis explicitly
Recover what you actually believed when you started — the specific reason this was going to work. If it was never written down, write it now; you cannot test a thesis you cannot recall.
Pro tip Sean's team wrote a one-pager investment memo before entering Luna, covering both why it was a good idea and why it could go horribly wrong.
- 2
Ask whether the thing you believed has changed
The question is not whether results are bad, but whether the underlying claim is now false. Facebook's price falling was not itself a change to what Sean believed about the company, so he bought more.
Watch out A bad outcome is not automatically a broken thesis. Sean's Luna loss was a downside scenario he had explicitly written down in advance.
- 3
Isolate genuinely new information
Distinguish new evidence from the same evidence hitting harder. When inflation spiked and people moved toward dollars rather than Bitcoin, that was a real result contradicting Sean's stated thesis, and he treated it as one.
Pro tip Ask what you would have predicted a year ago, then check the actual outcome against it.
- 4
Check whether ego is holding the position
Ask directly: am I holding on so I do not feel like I was wrong. Sean names this as one of the core questions and pairs it with whether you are willing to accept being wrong at all.
Watch out This is the question most people skip, and it is the one that produces stubborn locals.
- 5
Choose persist, tweak, or change your mind
Resolve to one of three outcomes: stay the course with small adjustments because you are right, make a real change because the signals warrant it, or exit. The test's value is forcing an explicit choice rather than drift.
Pro tip Sean's framing is that you are either being stubborn because you are right and just need to tweak, or banging your head against a wall — name which one out loud.
In the wild
With the stock getting crushed, Sean ran the test rather than reacting to price: what did I believe about this company, and has that changed. Nothing in the underlying view had changed, and no new contradicting information had arrived.
→ He added to the position — a persist decision reached by test rather than by stubbornness.
Sean had believed that as people noticed their cash losing purchasing power, they would migrate toward a monetary system without that problem. Inflation then hit its worst levels in decades and the response was conservative — people stuck with dollars, and the dollar actually strengthened as other currencies fell further.
→ He flagged it as the thing he has changed his opinion on most, and moved to reassessing whether the original logic was wrong rather than defending it.
Common mistakes
Reaching for a generic rule
Both the persistence story and the data-driven story are true in different contexts. Sean is explicit that nobody can give generic advice about when to do what — importing someone else's rule is how you get the wrong answer confidently.
Treating a bad outcome as proof the thesis was wrong
The Luna collapse followed exactly the downside path Sean's memo had described. Because the risk was priced in and sized for, the outcome did not invalidate the reasoning — conflating the two teaches you the wrong lesson.
Skipping the ego question
Without asking whether you are holding on purely to avoid feeling wrong, the other questions get answered in whatever way protects the position. This is the question that makes the test honest.
Is it for you?
Best for
Founders and investors holding a position or a product that is not working yet
Not ideal for
Anyone looking for a universal rule about when to quit — the test is explicitly context-dependent
From the transcript
“the hardest question as a Founder is to know am I being the right kind of stubborn right so am I being stubborn because I'm…”
“does the thing I believed about this has that changed do I have new information that makes me update my thinking here um do I…”
“nobody can give the generic advice about when to do what it's super it's like super context dependent super circumstance dependent”
From the episode
Sahil Bloom: Left Private Equity To Make Millions As A Content Creator And 3 Profitable Businesses To Start Now
Sahil Bloom