a student of good decision-making
Steve Jobs said there’s no difference between a great thinker and a great doer.
But most of the advice and “frameworks” are pointed at being a better doer. The OKRs, the project management tools, and the YouTube hustle culture. They all talk about doing more and getting things done. And with AI, the cost of taking an action will go down to almost zero, and if you’re just a doer and don’t know how to be a good thinker, then you’re handicapped.
After playing a lot of poker, and losing money on good bets, I wanted to systematically analyse the underlying mechanics of good thinking and decision-making. I see little difference between problem-solving and decision-making, so I will be using these words interchangeably.
Broadly speaking, we have four phases:
- Articulate the decision.
- Gather evidence for and against.
- Consequences and probability.
- Calibrate and iterate.
This structure should apply to almost every decision. Selecting a school for your kid, picking the right apartment to rent for the next 2 years, or choosing a co-founder.
1. Articulate the decision:
It might be a cliché, but a problem well stated is really half solved. A lot of people skip this step and assume they already know what they are working towards.
But when you actually sit down and write the problem statement well, you observe a lot of nuances that are very particular to your situation. Let’s say you have to decide where to live. If you state the problem as “what’s the best area to live?” vs “what’s the best area to live, where I can travel less to work and also have tight round-the-clock security?” the entire decision space changes.
So, do not neglect this step. Play with different framings and see what feels right for your situation. Phases 2, 3 and 4 are downstream from this phase.
My personal advice is to not use AI in this phase, as it’s heavily biased towards taking action to prove it’s worth the $20 subscription. It will pull you in the direction of what an average human would think, stripping away all the nuances of your situation.
2. Gather evidence for and against:
You have to start by really leaning into your instincts to gather solid evidence. After you have articulated your problem, most often your brain will give you ideas, and the chances that the ideas would be wrong are very high.
As Mark Pincus says in his book, Life at the Speed of Play, your instincts are right 95% of the time, but your ideas are wrong 75% of the time. You have to treat your instincts as a guide, but you have to refine your ideas by collecting the evidence for and against them.
Start by coming up with three to four ideas and then systematically collect evidence for and against them. You have to be creative and truth-seeking at the same time. You have to get creative with collecting evidence — run a cheap experiment, tell a story to capture the reactions, read books or watch YouTube videos, and talk to people. AI can be of tremendous help in this phase and become one of your many sources of evidence. It can take actions at almost zero cost and give you abstract evidence.
The harder part of the process is to be truth-seeking, especially collecting evidence against your already convinced heart and brain. You will naturally gravitate towards your own biases, and you have to resist that urge in order to seek the truth.
3. Consequences and probability:
Every good decision has an upside and a downside baked into it. Assess how likely those consequences actually are, and stop torturing yourself over worst-case scenarios that have a near-zero probability.
Let’s go back to our apartment example. Say you’ve narrowed it down to two gated communities. The first one is right next to your office, but the landlord is a complete asshole. The second one is 5 km away, but traffic might be a problem.
Weigh the upside and downside. You’ve spoken to the neighbours and the evidence suggests that the first landlord kept the deposits of the previous tenants. Losing your deposit is the bigger deal. Even at a 10% probability and with a lack of personal control, it’s a big no.
Now, by using the Google Maps “Leave by” feature for the second option, you’ve discovered that the chance you get stuck in traffic is lower if you leave early. It’s completely in your control, which makes the second apartment the much better choice of the two.
4. Calibrate and iterate:
You can avoid reality, but you cannot avoid the consequences of avoiding reality. — Ayn Rand
Poker taught me that the point of good decision-making is not a guaranteed good outcome, but better odds. With this solid backing, when you act, reality gives you quality information.
This new information can act as evidence to iterate towards a better outcome or to re-articulate the problem statement differently.
The process of making a good decision is continual; hence, it’s important to document the decision and the outcomes, so you can improve your mental model about a space and the world at large.
In conclusion
Being a doer and a thinker is a continuum. The two don’t exist in silos, and systematically analysing them can help you improve them further.
In today’s hustle culture, action gets most of the applause and good decision-making gets very little. Yet good decision-making shapes everything in our lives. A well-made car, a sturdy flyover, a product that just works. Nobody looks at a flyover and thinks about the decisions behind it, and that’s the point. Good decisions look simple and obvious, because all the complex thinking happened before anyone saw them.
Bad decisions are the opposite. They’re muddy, noisy and half-baked, and everyone notices.
That’s why it’s important to be a student of good decision-making, always trying to understand how to get better at it.