The Three-Question Opportunity Filter
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 6
- Confidence
- —
Before saying yes to any new project, side hustle or content bet, Dickie Bush runs it through three questions. First, is there no way this is a failure — is the worst-case outcome still a positive one, such as learning a skill or understanding the material better? Second, is there exponential upside if it works and you stick with it long term? Third, can you actually stick with it long enough to see the benefits of compounding, which he benchmarks at around two years. A yes to all three is a potent combination and a candidate for something you could do for the rest of your life. The filter also biases you toward formats that package work you are already doing, so the marginal time cost stays low and the two-year commitment stays realistic.
Origin
Extracted from Deep Dive with Ali Abdaal
How to run it
- 1
Define the opportunity as a concrete, repeatable format
Not 'build an audience' but 'write one newsletter a week' or 'publish one video a week'. The filter only works on something specific enough to commit to.
- 2
Question one: is there no way this is a failure?
Identify the worst case. If the downside of writing every day for a month is that you understand the podcasts you listen to better, the downside is zero. Require a worst-case scenario that is still a positive outcome.
Pro tip Write the worst case down explicitly rather than assuming it is fine.
- 3
Question two: is there exponential upside?
Ask whether persistence could produce a non-linear result — audience, distribution, compounding skill, a product. Linear-payoff work fails this test even when it is comfortable.
- 4
Question three: could you stick with it for two years?
Project the format forward two years of consistent output. If the honest answer is no, do not start — the compounding never arrives.
Watch out A format that requires 20 hours of fresh research per piece will not survive contact with a full-time job.
- 5
Reduce the marginal cost by packaging work you already do
Design the output so it distils something you are doing anyway — the books you read, the experiments you run, the job you already have. This is what makes the two-year answer a yes.
Pro tip Tap pre-existing expertise rather than a topic that is merely trending.
- 6
Commit to a fixed rep count and start
Convert the yes into a bounded challenge — 30 days of writing, 52 weekly newsletters, one video a week — so the decision is made once and execution becomes mechanical.
Pro tip Shorter challenges accelerate the feedback loop when the long-form format is not working.
In the wild
Nine months into a weekly newsletter with about 100 subscribers, Dickie switched to writing on Twitter for 30 days specifically to accelerate his feedback loop. The downside was zero: worst case he would simply understand the podcasts he was listening to better. On day 28 a thread about biology went viral, taking his newsletter from roughly 100 to 500 subscribers overnight and unlocking the audience that made the first products possible.
→ A no-downside experiment produced the exponential upside the filter predicts.
Applied to Ali's YouTube channel, the worst case was better public speaking, more reading and more explored curiosity — no downside. The upside proved exponential as the channel grew to millions of subscribers. And one video a week was a cadence he could plausibly sustain for two years, especially because the early videos summarised study techniques he had already learned rather than new research.
→ All three questions returned yes, and the channel compounded for a decade.
Common mistakes
Picking a format you cannot sustain for two years
A crypto channel requiring 20 hours of research per video multiplies the effort per unit of output. It passes the upside test and fails the stickability test, so it dies before compounding starts.
Treating 'I might fail' as a real downside
If the process itself teaches you something, failure is not a downside. Mislabelling a no-downside experiment as risky is what keeps people from ever starting.
From the transcript
“so I like to think of when you're evaluating a new opportunity you should think about three questions before you say yes to it”
“so you should have a worst case scenario that is still a positive outcome second you need exponential upside if you are able to stick…”
“and then third you need to be able to stick with it can I stick with this long enough to see the benefits of compounding”
From the episode
How to Build a Business that Lets you Quit your Job - Dickie Bush
Dickie Bush