Second-Order Thinking

February 24, 2020

One thing is for sure, and that is that decisions we make will always have more consequences than at first appearance. The world is not a simple place, and any changes will not only have the initial consequences (that we probably intend) but also affect other things both now and far into the future. This can be both for good and bad, but either way we rarely consider them.

First, an example of first and second order thinking. Imagine you are deciding whether or not to go to the gym:

  1. First order thinking says that it will make you tired and you will be sore afterwards. This logic suggests you shouldn’t go.
  2. Second order thinking says that following consequences to that decision might be feeling better, looking better, and having more energy in the long run.

(Disclaimer: It’s an example, you don’t have to go to the gym if you don’t want to 😉).

Notice how these consequences are opposites. This highlights why you can easily fall into traps by only considering the first order. Also consider the alternative! It works vice versa too, sometimes first order thinking might make something sound like a great idea, but the bigger picture longer term consequences could be hugely bad!

First-order thinking is simplistic and linear. Most people would come up with the same decision based on this order. Second-order thinking is complex. You need to consider the bigger picture, wider scope, systems and all people involved, time involved etc. This is where long lasting solutions, creativity, innovation and above average outcomes appear.

Second-order thinking is hard to do, not only because of the complexity but because of the extra workload. It requires sufficient more effort to consider the second-order in complex contexts (like business), but sometimes it’s important to do.

Who should make use of second order thinking?

Everyone. We all make decisions in our lives and work, and the more often we add n-order thinking to our problem solving the more long term successful outcomes we will achieve.

However, I’ll talk here specifically about becoming a more senior/leading engineer. There are two scales that increase as your career develops in that direction: Scope of influence and timescale impact. A junior engineer might only work on specific tasks and only thinking ahead a couple of weeks regarding the current tasks they are on. A super senior engineer might be considering consequences of decisions across our whole stack and how it will impact us over the next year’s worth of not only development but also sales/revenue/strategic decisions.

To become a/better tech lead, you should practice sliding upwards on both of those scales. Second-order thinking is a principle you should keep in mind to help you explore those bigger scopes and timelines.

Tech lead example: A new product feature

Imagine you have a request to build a new product feature. The only solution you have so far adds 200ms to the payment dialogs load time. First order thinking would suggest that this is a bad idea so we shouldn’t ship it. But now start to consider the second order:

  • Who will use this feature? Will it be for a big significant partner that means revenue/investment for us?
  • Who else will it impact in the company? Will sales close some open deals because of it? Will it affect engineering guidelines which set a minimum load time? Who owns those guidelines?
  • Will the raised awareness for load time be a great learning opportunity for the company? Could it lead to a project to reduce load time massively so we have more head room in case of these type of changes? Might we develop an innovative way to handle reducing load times that we could share with the world?

Despite the first order consequence sounding far from ideal, answers to the above will help you understand whether second order consequences make it worth it. If you can be the person to lead the exploration into that kind of thinking, you’ll help everyone along the way and ensure things your team does has the bigger picture and long term in mind.

Ok, but how do you do it in practice?

1. Run through some go-to questions

  • “And then..?” - Consider what happens after a first order consequence. Keep asking until you uncover the bigger picture, similar to when a child keeps asking “why.. but why…. but why…” (kids are great to learn from)
  • “What would the opposite conclusion mean?” - Consider the opposite of a first order conclusion. For example, if you think an idea is good, practice being a devils advocate. This forces you out of the box and into the realm of the second order.
  • What is the range our future outcomes? Which are likely to occur? What probability do i think i’m right?

2. The timescale evaluation

Ask yourself: What will the consequences be in x weeks, x months, x years from now? Do things change as time goes on, do the consequences get more positive or more negative? Are any of them significant?

3. Design thinking

Design thinking is a method that is used in product development and it makes use of higher order thinking to step back into a problem space to define the challenges that exist in the whole context of the system and everyone involved. See Double Diamond framework.

4. Lateral thinking

This is basically “outside the box” thinking. There are techniques you can try, here’s some.

5. Map your consequences

When you are exploring solutions, use a table to map out consequences as 1st, 2nd, 3rd etc. order columns with the consequences under each. This helps you keep track and visualise good or bad outcomes in different orders, to help make a balanced decision.

Final words

Second-order thinking is something you’ll have already done in your life at various points, but it’s not something we often consciously consider. Being aware that initial conclusions are not a true reflection of the bigger impacts of decisions is important in helping us make informed, sustainable and long term decisions. By introducing a few practices listed above into your thinking process, you can help yourself explore second order thinking purposefully.


comments powered by Disqus