Decide from principles, not either/or thinking

The Overlap #13

Welcome to The Overlap, a biweekly newsletter that explores the relationship between product & organization design. It comes out every other Wednesday morning.

This one’s on decisions, either/or thinking, and first principles. Here’s the takeaway: when making a consequential decision with your team, reason from principles first before reasoning about the options.

Let me guess the way you make tough decisions.

When presented with a tricky choice between A or B, I’ll guess that you:

  1. List out all the pros and cons of A.

  2. List all the pros and cons of B.

  3. Decide.

The problem with making a pros and cons list between two options is that it assumes that they are your only options. Your conception of the problem immediately becomes an either/or choice. Hire the candidate or don’t hire the candidate. Vote for Biden or vote for Trump. Build the technology or use a third-party solution. 

Looking at a decision through an either/or lens is helpful in certain situations. However, it closes you off from other possibilities. Decisions are rarely a binary choice, and there is always a Decision C (maybe even a D, E, and F). So how do you find it?

Decide from principles instead of options

When teams are presented with a choice between A or B, here’s an approach I’ve found helpful:

  1. Revisit our principles.

  2. Discuss which principles are most important to our team. Rank them in order of most important to least important. 

  3. Look at the currently proposed options.

  4. Ask: what are other options we’re not seeing? Write them down.

  5. Decide, based on your most important principles.

This is similar to the way Shreyas Doshi makes decisions with his teams:

Define your principles first!

If your team doesn’t have principles, define them with your team first before running through this framework.

To define your principles, understand this: Principles are not values. 

In my mind:

  • Values are things we believe to be good.

  • Principles are universally useful patterns that guide action regardless of context.

I like the way Simon Wardley explains the difference between principles and values:

X: What do you mean by principles?

Simon: Universally useful patterns that apply regardless of context

X: Are these the same as values?

Simon: No. Values are things you believe in rather that universally useful principles of action. When it comes to values some are mutually exclusive and some non-exclusive.

By this definition, some design principles are just values. I’d argue that Dieter Rams’ “ten principles for good design” are all values instead of principles.

Of course, good design is aesthetic. Of course, good design is thorough to the last detail. And of course, good design is as little design as possible.

The problem with this list is it doesn’t help us make tradeoffs. Sometimes we have to de-prioritize aesthetic to prioritize thorough to the last detail—just look at Roam’s aesthetic. Other times we have to prioritize honesty, which exposes the fact that we’re not so environmentally-friendly.

Yes, this list presents good defaults. But your team needs to turn them into principles. You can use even/over statements to do that:

Even over statements are written like this:

A good thing even over another good thing

Value: Good design is environmentally friendly

Principle: Environmentally friendly even over aesthetic

Again, values are things we believe to be good, while principles are universally useful patterns that produce action.

Now, let’s take a step back and clarify what first principles mean. Understanding first principles can break you and your team out of the either/or mindset.

What are first principles?

A first principle is a basic assumption that cannot be deduced from any other proposition or assumption. First principles stand alone.

We can surface a first principle in any situation. Let’s surface the first principle behind what’s wrong with “good design is thorough down to the last detail”, assuming “everything is situational” is a first principle.

Again, principles are universally useful patterns that produce action regardless of context. For “good design is thorough down to the last detail” to be a principle, it needs to apply universally, which it does not. Here are three examples of when good design isn’t exactly thorough:

  1. Sketching a paper wireframe of a checkout flow to understand the different ways of designing the flow.

  2. Creating a prototype of one feature while holding off on the seven other features on your backlog, because these seven other features depend on this feature being a thing users will spend time on.

  3. Andy Matuschak’s Working Notes. Andy doesn’t have every last detail of every note filled out, and that’s the reason it’s so great: his notes are constantly adding, revising, and evolving itself. 

Since “good design is thorough down to the last detail” does not apply universally, it is not a principle. It’s a value. But what can we learn from this? And what is the first principle behind this learning? 

Again, a first principle is a basic assumption that cannot be deduced from any other proposition or assumption. Therefore, we can use inductive reasoning—just asking “why?”—to get at a first principle.

Let’s question the value of “good design is thorough to the last detail,” using Andy Matuschak’s Working Notes as an example.

  • This design is unique. It’s rare for people to expose their personal research notes to the world.

  • This design isn’t exactly thorough down to the last detail. Some notes are fleshed out, while others are intentionally left blank.

  • However, I’d still consider this design good.

  • Perhaps good design isn’t always thorough to the last detail. Instead, good design knows when to be thorough and when not to be.

  • Why is this the case? 

  • Because everything is situational.

  • Why is everything situational? 

  • Because... uh... I don’t know, it’s a first principle!

By questioning “good design is thorough to the last detail”, we found our first principle: everything is situational.

Why are first principles useful?

There are two reasons first principles are useful.

First, they prevent the status quo bias: our tendency to assume that the way we do things is the best way.

Elon Musk makes this point:

I think people’s thinking process is too bound by convention or analogy to prior experiences. People rarely try to think of something on a first principles basis. They’ll say, “We’ll do that because it’s always been done that way.” Or they’ll not do it because “Well, nobody’s ever done that, so it must not be good. But that’s just a ridiculous way to think. You have to build up the reasoning from the ground up—“from first principles” is the phrase that’s used in physics. You look at the fundamentals and construct your reasoning from that, and then you see if you have a conclusion that works or doesn’t work, and it may or may not be different from what people have done in the past.

Second, deciding from first principles opens new possibilities.

I’ll use a real example that you may relate to. When the pandemic was breaking out in NYC in March, our company faced a crucial decision: should we encourage everyone to work remotely?

To think through this, we had a conversation around seven questions. These questions helped us think about this from first principles.

1. Why do we think going remote is a good idea?

  • Working from home decreases the chances of the coronavirus spreading. 

  • Some team members feel safer working from home.

  • Other notable companies are doing it (Facebook, Twitter, Coinbase).

2. How do we know this is true? 

  • Well, we know that the coronavirus spreads rapidly. If folks stayed at home, there would be fewer people gathered, and the fewer people are gathered, the less of a chance that the virus will spread. Therefore, working from home decreases the chances of coronavirus spreading. 

  • If I commute to the office on a crowded subway, I wouldn’t feel safe. I’m going to assume that others might feel the same way and that I’m not alone in feeling safe by being able to work from home. Therefore, working from home decreases the chances of coronavirus spreading. 

  • Coinbase published an essay on how they were handling the pandemic. They seemed to be taking it seriously, so we consider taking it seriously too.

3. What if we thought that going remote isn’t a good idea? 

  • It’s tough to serendipitously collaborate in remote work. Working on a whiteboard spontaneously with a teammate is a lot easier if my teammate and I are in the same physical space.

  • Working from home makes it hard for folks to learn from others. Entry-level employees often learn through pairing—sitting side by side in front of a monitor and working on a problem together.

  • Working from home feels isolating. Feeling connected and a part of a team leads to feeling satisfied at work.

4. What might others think? How could I prove myself wrong?

  • Sean Blanda, a writer I trust, wrote about why working from home sucks.

  • Look at my own experience of when working remotely sucked: the office doesn’t suffer from unexpectedly dropping from a Zoom meeting, dogs barking, people forgetting to be on mute.

  • I enjoy my job because it’s all about thinking with others. In remote work, there’s more friction to get into a space where my team and I can think together, even with the advent of collaboration tools.

5. What are the costs of us being wrong? Are these costs reversible?

  • We’re increasing the chances of COVID spreading, which increases the chances of folks getting sick and/or at risk of death.

  • Death isn’t reversible.

6. What conclusions can we draw from these questions?

  • Going remote is worth trying. We’ll retro how it’s going in a month and decide to keep it or go back to being in an office. (Obviously, we stuck with remote work.)

  • There’s a better way to frame the problem. It’s not “remote vs in-person.” It’s “How can we use this moment to design the company we want to work at?”

7. Were we correct?

  • Yes! We’ve decided to stick with remote work until everyone is vaccinated. Most team members enjoy the flexibility of remote work. People get more focus time in their day, too. And it’s great that we’re doing what we can to maintain the curve. We still want the option to work together in person, but we also want remote work to be an option whether we go back to being in an office or not.

From The Overlap #8:

Through these experiments, we aren’t just trying to adapt our culture to an all-remote world. We’re co-designing a new version of our culture. Two very different approaches.

Reasoning from principles is slower, but pays off

When you encounter a consequential decision with your team, resist the urge to immediately jump into the pros and cons. Reason from principles first before reasoning about the options. It takes some getting used to, but it pays off. 

What I’m Reading

1. First Principles: The Building Blocks of True Knowledge

Great overview of applying first principles to your day to day decisions. 

2. Eigenquestions: The Art of Framing Problems

Making good decisions starts with framing the problem accurately and artfully. Problem framing is a skill we should take seriously:

3. What’s the fuss about roles in horizontal teams and organizations?

Remember this idea from The Overlap #11?:

To distribute authority without going full out on self-management, learn to separate role hierarchy from positional hierarchy. 

Samantha Slade wrote the most succinct guide to getting started with structuring your org based on roles (rather than titles). Give that a read.

4. US Government Org Chart

I think this is my favorite visualization of the US government’s org chart. If you’re ever looking to buy a map, check him out. Not an affiliate partner. Just a fan.

See you in two weeks,

Thumbnail image from this block.