We are judged by our decisions. If they’re bad, the manager takes the blame.
When I was an engineer, things were much easier. I could run tests, keep good test coverage, and, in general, sleep better at night.
As managers, we can’t test our decisions the same way. But there is one thing we can do: we can use mental models to help us prevent huge mistakes.
I call these thinking power-ups. These help with changing perspectives to help us make solid decisions and find novel solutions to our problems.
In today’s article, we are going to cover 7 mental models I found especially useful for engineering managers:
Inversion
Interia
Second-order Thinking
Entropy
Hanlon’s Razor
Bottlenecks
Activation Energy
Today’s article is a guest post by , the author of - one of my favorite newsletters!
1. Inversion
What if you think about how to make your team demotivated? — It’s important to understand what you should avoid.
Invert the problem to look at the opposite perspective. You can list all the things that should be avoided.
Usefulness: It helps with avoiding stupidity, which is more important than seeking brilliance.
In Practice:
Instead of thinking about how to use our development budget, list things you don’t want to use it for.
Instead of thinking about the perfect engineer that fits your team, list what traits make the worst team member, and what should be a red flag during the recruitment process.
Instead of thinking about how to cooperate with other teams better, focus on things that should be done to make it worse.
2. Inertia
Do you keep executing the same old processes even if they do not serve the purpose?
Inertia keeps things moving — if not disturbed by friction or any external force, they will keep moving, that's Newton's First Law of Motion. It's a tendency to resist changes in a state of motion. We can use it as an analogy for our actions while leading developers.
Usefulness: It pushes you to reconsider all actions that you are following because you are used to them.
In Practice:
You don’t lead team meetings because you didn’t start doing these. You maintain the status quo.
You keep doing the daily standup the same way even if the format doesn’t work as it was a few months ago.
You don’t try to play with AI tools to check how productive you can get because you like your old code editor.
You don’t ask your manager how to get promoted, as you like how comfortable things are now.
3. Second-order Thinking
How often do you consider more than the immediate results of your actions?
Second-order thinking is a mental model where you consider more than just one immediate result of your decisions. As its name states — you also look at potential future results (second-order consequences).
To apply in practice you can ask “And then what?” or consider the consequences of your action over a larger time-span. What will happen with this: In 10 days? In 10 weeks? In 10 months?
Usefulness: It makes you think long-term.
In Practice:
You need to push the feature to show it during the demo. The decision is to do it fast, without tests:
First-order consequences: it works during the demo.
Second-order consequences: it breaks the whole app as it was an unstable piece of code just for the demo.
You stop doing team meetings because you don’t have time for it:
First-order consequences: people are happy as they have more time to work, and you are even more happy as you don’t need to bother leading them.
Second-order consequences: engineers don’t have a shared space for knowledge exchange, and their morale deteriorates over time.
You change your CI/CD suite to use a better solution:
In 10 minutes: everyone sees how painful the migration will be.
In 10 days: still a mess, as migrations are on the way, but part of the apps are already using the new flow.
In 10 months: no one even remembers the old tool we had.
4. Entropy
Do you see how your code and processes degrade over time?
Things will eventually degrade without taking action. Entropy guarantees this. Entropy is the measure of disorder. The more disorganised things are, the higher the entropy. We need to put energy into making sure we will keep operating at the same quality.
Usefulness: It keeps you reminded that the quality will degrade over time without putting in effort.
In Practice:
Your codebase is going to get worse and worse without proper maintenance, as its complexity increases.
People need reminders, need reinforcement — Imagine training for your team in security measures or any kind of best practices, it’s all by the book when it’s fresh but then without reinforcement security checks might get less and less detailed.
Best practices used in the projects might not be the best practices in a month or two, be aware of “best” naming here — “best when created” but not necessarily right now.
5. Hanlon's Razor
Does your team assume that business people don't like them?
“Never attribute to malice that which is adequately explained by stupidity.” — In other words, assume good intentions.
When you don’t have the mindset of assuming good intentions, it just makes operating difficult if you need to take into consideration humans who wish you to fail.
Usefulness: It frees your mental capacity from trying to understand why others are threatening you.
In Practice:
The business team promised a feature that is impossible to deliver — do you assume they want you to fail? Or maybe they just don’t understand the app?
The company made mass layoffs and chose you — do you think it’s because they don’t like you? It could have been a reaction to market changes or simply bad management. There’s no point in taking it personally.
6. Bottlenecks
Do you feel like without you the company or team will break apart?
Where does it all fail? Usually in the most vulnerable part — the bottleneck.
The necks of the bottles make them easier to drink from and pour out of, but they limit the flow. This concept has been adopted when we are speaking about components that are limiting factors for the systems.
Usefulness: It reminds you that a chain is only as strong as its weakest link.
In Practice:
There might be one great engineer who holds all the knowledge — if they leave, it could significantly slow down the work.
A supporting team in the organisation, such as Platform or IT support, may become a bottleneck, slowing down the entire organisation.
You might be a bottleneck — your capacity is limited, and if everything in your team goes through you, you become the limiting factor.
A slow element in your CI/CD process may reduce your team’s ability to deliver changes quickly.
7. Activation Energy
Do you repeat that you will do something but you never actually start?
Activation energy is a concept drawn from chemistry, where a certain amount of energy is needed to start a reaction.
Usefulness: It forces you to start but also reminds you that the energy is needed to maintain the new process.
In Practice:
The onboarding process is costly and takes a lot of time to get new engineers up to speed. When activation energy is high, people are reluctant to hire more. To help with this, use checklists and trained mentors. Make sure to improve the process incrementally with each new engineer onboarded.
When introducing new processes, such as team meetings, remember that just starting is not enough — activation energy must be sufficient to sustain them over time. Decide on speakers in advance — the further ahead, the better.
Bonus: The Perspective Training
Are you overwhelmed by difficult situations?
To understand reality our own view is sometimes not enough. Looking at problems from the perspective of different people helps us to understand them better.
I worked as a manager at my company which was preparing for mass layoffs. We got professional training sessions to prepare for the unfortunate event. These sessions were helpful. However, what mattered most to me was “the perspective training”:
I watched Up in the Air, starring George Clooney. The protagonist travels around the US, firing people because their bosses are too afraid to do it themselves. It’s not a masterpiece, but it shows the worst-case perspective. In the movie, all the people losing their life roles are devastated, broken by the process, and have nowhere to go. I knew it would be different for me — we were letting good people go who would soon find good roles elsewhere. This helped me to keep me calm during the layoff, and looking back, all of them did find new jobs.
Get perspective of the future — Will it matter in 10 days, 10 weeks, 10 months, maybe 10 years from now?
Summary
There is no magic pill that will make your manager's career smooth sailing, but we can help with that by understanding what makes our decision better.
This set of proven mental models will help, as long as you remember to apply them.
Thanks for reading,
— Michał
What I enjoyed reading this week
How to lose friends and alienate people (the story of WeWork and Adam Neumann)
