The Four P's: Potential, Probability, Passion, Prowess
Score any idea on four factors — and always compute potential before probability.
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 95%
A four-column spreadsheet for evaluating any startup idea, new product, or feature. Each factor is scored 0-10. The critical sequencing rule is that you must assess Potential (magnitude of the outcome) before Probability (odds of success) — because reversing the order causes you to filter out low-odds, enormous-payoff ideas before you ever compute their expected value.
Origin
Dharmesh Shah's own idea-evaluation framework, built out of his preference for quantifiable factors he can multiply. Expected value is standard decision theory; the sequencing insight and the four-P structure are his.
Core principles
- 01Score everything 0-10 so you can multiply, rather than using letter grades.
- 02Potential first, probability second — reversing them applies a filter that kills asymmetric bets.
- 03A 10% chance at a $100M outcome has a higher expected value than a 50% chance at a $10M outcome. Do the maths before you trust your gut.
- 04Passion is overrated and ambiguous as a starting condition — most successful founders became passionate after digging into the space, they did not start there.
- 05Prowess is your unfair advantage — existing code, market access, distribution — and it directly feeds back into your probability of success.
- 06No single factor decides. The discipline of walking the four columns is the value, not the final number.
How to run it
- 1
Score Potential
If this thing worked, how big could it be? Score 0-10 on whatever magnitude matters to you — revenue, market cap, impact. Do this before anything else.
- 2
Score Probability and compute expected value
Only now estimate the chance of success. Build a simple decision tree if useful (50% chance of $10M = $5M expected value; 10% chance of $100M = $10M expected value) and compare across ideas.
Pro tip Adjust for your stage of life — how much variance you can personally absorb changes which expected value you should chase.
Watch out This is the step people do first, and doing it first is the single most common mistake. A '1 in 10 chance' filter fires before you ever notice the payoff is $10 billion.
- 3
Score Passion or Proximity
Do you actually care about this problem? Is it something you can work on for two, ten, or twenty years? Proximity — genuine closeness to the problem — is the more reliable version of this factor.
Pro tip It is fine to become passionate about a space after you enter it. Do not wait for a calling.
- 4
Score Prowess
Ask: why me? What asset gives you an unfair advantage — reusable code, an existing customer base, distribution, domain knowledge? Why would your chance of success be higher than anyone else's?
Pro tip Prowess is not an independent column — it should raise or lower your Probability score.
- 5
Stack-rank the factors instead of forcing exact weights
You will not be able to assign precise weights, and you should not pretend to. Rank which factors matter most for this decision, then let the human make the call with much better information.
Watch out Do not let the spreadsheet make the decision. Data informs; people decide.
In the wild
Lenny cites one of his favourite investments: the Zip founders were not remotely passionate about procurement when they entered the space. They saw a large opportunity, believed they could build a much better product, and developed genuine passion for it once they started working on the problem.
→ It validates Shah's ordering: potential and prowess drove the decision, and passion followed — rather than being a precondition.
Applying the same logic to HubSpot's founding: SMB software had exactly one global success story (Intuit), so probability looked terrible and capital was nearly impossible to raise. But the potential (millions of customers, no revenue concentration, short feedback loops) and the founders' prowess (Shah on product/engineering, Halligan on sales/marketing, both burned by enterprise) justified the bet.
→ HubSpot went public at roughly $1B in market cap and now sits around $30B, having never pivoted away from SMB in 18 years.
Common mistakes
Filtering on probability first
Applying a mental odds filter before you have sized the prize causes you to discard exactly the asymmetric, high-expected-value ideas that produce outsized outcomes.
Waiting to find your passion before you start
Truly knowing your calling is rare, especially for first-time founders. Passion is frequently an output of doing the work, not an input to choosing it.
Optimising a single column
Picking whichever idea has the highest expected value while ignoring prowess and proximity produces ideas you have no right to win and no stamina to pursue.
Is it for you?
Best for
Founders choosing between startup ideas, and product leaders deciding which new product or feature bet to fund.
Not ideal for
Small, reversible, low-consequence decisions where the analysis costs more calories than the decision is worth.
From the transcript
“so that's the potential thing number two is probability of success okay and now I'm going to pause here and tell you the most common…”
“so start with the kind of potential outcome then look at probility not the other way around”
“the third thing is is uh what I think of is either passion or proximity”
“it's like prowess like do you have some assets some something that makes you uniquely positioned”
“why me like why why me why my company why would we succeed at this”
From the episode
Zigging vs. zagging: How HubSpot built a $30B company
Dharmesh Shah (co-founder/CTO)