My Decision Making Process

Over the years, I’ve developed a decision making process based on researching psychology, decision-making literature, and my own personal experience. Several people have commented that my procedure for making decisions often leads to very good choices very quickly, and in some cases it’s helped them make a decision in minutes or hours that they were agonizing over for weeks. The core idea is to figure out which pieces of information are essential at any moment in time, and focus your effort on those.

The Process

  1. Have a list of concrete alternatives.
  2. Pick one alternative as the default, then compare alternatives two at a time.
  3. Form a tree of decisive considerations.
  4. Focus on getting the information needed to solve the decisive considerations in order, until a clear winner emerges.
  5. Replace your default with whichever alternative was better, then compare it to the next option.

Example: Choosing Between Jobs

For step (1), alternatives like “work at Google” or “don’t work at Google” are bad, options like “work as a software engineer at Google”, “work as a data scientist at Sciencely”, or “continue searching for jobs” are much better.

Then you’d pick one somewhat arbitrarily as the default–say “continue searching for jobs”, and compare it to one other option. It’s important to take just two options at a time, because this makes comparison dramatically easier.

To form a tree of decisive considerations, think about things that would be definite deal breakers, or definite acceptance criteria. If Sciencely requires you to use Windows, and you absolutely can’t stand working on anything except ArchLinux, that’s probably not the job for you. A tree of considerations might look like this:

  1. If the job requires Windows, definitely no
  2. Otherwise, if the job requires relocating, definitely no
  3. Otherwise, if the job pays below minimum salary requirements, definitely no
  4. Otherwise, if the job looks like it will help you learn crucial skills substantially faster, definitely yes

…and so on. You don’t have to know what the answers to the decisive considerations are at the beginning, the point is that laying this out will help you figure out what to do next. If you have a tradeoff–for example, you’d be willing to use Windows if the job paid twice as much as anything else–then you can reframe condition 1 as “If the job requires Windows and pays less than twice as much as the next best job, then definitely no.”

Once you have this list, focus on figuring out the information you need to settle the decisive considerations, in order. You’d ask about Windows first, and don’t bother exploring the rest of the details. You go on until you’ve either hit a decision, or exhausted the tree. In case you exhaust the tree, add more branches.

While this process looks fairly long written out, it’s often very quick because there’s a clear decisive consideration. For example, I chose which college to attend in about 2 minutes doing this–while there was a huge flood of possible information I could have looked at, it became very clear that there was just one decisive consideration that ruled out almost every option.

Why does this work?

Having multiple alternatives avoids a very common decision-making mistake, which is decisions of the form “take the job or don’t take the job.” One is much more concrete, easier to visualize, and easier to get information on, so people will almost always end up taking the concrete alternative, and usually end up regretting it.

The second benefit of this approach is that it accounts for the fact that humans have limited working memory, so your mental performance tends to drop off rapidly the more items you hold in you head. By doing things like comparing only two alternatives at a time, the number of things you have to pay attention to is greatly reduced.

Another important factor is that the tree of decisive considerations reduces what can seem like very complex values problems to relatively objective questions which can be answered with information. This makes it very easy to narrow in on the actions you need to take.

Finally, this process makes it easier to change your mind if some of your information was wrong. If you were originally going to pick Google over Sciencely because of the Windows issue, but then you find out that they’re willing to change what systems they use, that can very easily update your decision. On the other hand, if you made a decision in a more intuitive way by taking in lots of information and going with whatever feels best, it can be very hard to incorporate new data.

Categories: Focus, Productivity


  • u

    Hmm. So this is definitely interesting, but don’t you lose a lot of information by expressing your preferences as Decisive Considerations? In real life I think most questions are really about complicated multi-factor tradeoffs that you couldn’t express in this “decisive considerations” way. I could still see the method being useful — maybe what it loses in accuracy is sometimes made up for by speed of decision-making — but it seems to me it has a lot of drawbacks as well as virtues. And I’d be keen to hear more what you think of the drawbacks.

  • Satvik

    Can you give an example? I very rarely run into choices that don’t have decisive considerations, though sometimes figuring out what they are or getting the information needed to solve them ends up being a lot of work.

  • u

    mmm… this will be really hard to describe because I don’t have the vocabulary for it but I guess what I’m saying is that I think my decisions are normally more like “if the job requires Windows definitely no, unless it pays double in which case maybe, but I’ll accept $1k less for every mile closer to home, and I’ll accept $20k less if I get to work with close friends, and “. So it’s tradeoffs all the way down, especially because any super-decisive considerations (e.g. “I’ll never work in [industry X],” “I need to be in [place] because my partner lives there”) have already determined which jobs I’m even looking at.

    I can see why the decisive-consideration method might be useful — perhaps the extra speed and decided-ness of the decision-making outweighs losses in accuracy — but now it’s more of a pros-and-cons situation than you described. Does that make sense or am I totally missing your point? Or does your brain/situation work differently from mine?

  • Satvik Beri

    Thanks for the example. I don’t really see it as a trade-off between accuracy and speed. I think there are two big differences: first, when I only have concrete alternatives, a lot of possible factors fall away, because only differences between the two alternatives can matter. Second, I tend to place a pretty high priority on having one main reason (or a small number of reasons if necessary) for any decision, because that yields much better feedback on whether the decision was right or wrong–this is more of a trade-off of short-term accuracy for longer term improved decision making.

Leave a Reply

Your email address will not be published. Required fields are marked *