Two helpful razors to sharpen your mind with

Not the ones you'd like to harm anyone with, but rather helpful tools to make sense of challenging or puzzling situations

Today’s article will introduce two of the most helpful razors I know of, following a recent article where I introduced two helpful paradoxes that every engineering leader should know about.

Two helpful paradoxes, and moving home

By the way, in that same article I announced that I'm planning to move over to Ghost and away from Substack. Your support will be greatly appreciated to ensure all the content stays open forever and not behind a paywall as that move happens. Thanks for considering upgrading to a paid tier and showing your support for the publication.

Before you become worried that I might be suggesting ways to harm other people or even yourself, let me reassure you: chatbots and political leaders are already doing a great job at that. There’s nothing I can meaningfully contribute to that conversation.

Photo by Stefan Schauberger on Unsplash

Let’s start with clarifying what we’re actually looking at here.

Two Helpful Razors

We’re not talking about physical razors, the ones commonly used to shave and occasionally to perform more questionable activities. I’m talking about philosophical razors: rules of thumb, or principles, that allow a person to quickly “shave off” unlikely explanations for a fact or phenomenon, focusing instead on the more likely one.

These aren’t precise, scientifically proven tools, but rather heuristics derived from empirical observations of the two persons after which they’re named. I’ve personally found them to be enormously helpful on countless occasions, so do the many people I had the opportunity to share them with.

I believe they should be in the toolbox of every person who’s ever faced important decisions requiring the interpretation of facts or behaviours. If that sounds strangely familiar, it’s because it is. As leaders, you’re constantly dealing with similar situations.

These are Occam’s and Hanlon’s razors. While the first one seems to be the most well known, in my experience, the second is by far my personal favourite.

I’ll save the best for last, as someone once said.

Occam’s Razor

This one is often used and sometimes abused. Its original formulation, attributed to William of Ockham, was in the following form.

Entities must not be multiplied beyond necessity

This has evolved over time into the most common or popular version, often phrased as such.

Of two competing theories, the simpler explanation of an entity is to be preferred.

This sounds interesting, but before we move forward, let’s clarify something important. The two competing theories mentioned above are not merely crazy ideas made up by someone off the street. Rather, they’re supposed to be hypotheses that both have equivalent explanatory powers to describe the observable phenomenon or situation. This means both theories should have some form of validity before the Occam’s razor can apply.

This rules out using Occam’s razor to disprove flat earth-related theories, as they don’t satisfy the need to be coherent and validated before being subjected to the simplicity tradeoff.

I brought up simplicity because that’s the essence of Occam’s razor. But misusing it can lead to oversimplifications and manipulations.

Which reminds us of the famous quote attributed to Einstein.

Everything should be kept as simple as possible, but not simpler.

I know two common situations in which this tool can help engineering leaders: one is dealing with people, and the other is making technical decisions.

How often have you found yourself suddenly observing a team member's performance change, usually for the worse?

If you’ve been running teams for more than 500 milliseconds, you’ve likely been in that situation. When this happens, we often fall for the temptation of coming up with very articulate explanations for the observed change.

Maybe they’ve contracted dementia and are suddenly losing all the skills they used to master. Maybe they have fallen in love with someone who works at a competitor and are silently sabotaging your company from the inside. Or maybe they have always been lousy performers and have somewhat managed to cheat their way up until recently, making you realise you have not been paying attention for months.

Occam’s razor won’t tell you what the real explanation is.

In all these cases, I tell people to simply ask the person in question to help you understand what caused your change in perception. Yes, it's your perception that has changed, and you still need to validate that it's grounded in facts rather than speculations.

You should get into that conversation with as little bias as possible, so among all the crazy explanations you might want to bet on before discovering the truth, picking the simplest one that can reasonably explain what is happening is a good heuristic. If anything, you’ll end up wasting fewer mental CPU cycles on hallucinations¹.

My general recommendation when dealing with people’s behaviour is to forget Occam’s razor, as it encourages the annoying tendency we have to make up explanations for their behaviour, and just speak with them.

But if you are in a situation where that’s not possible, then use Occam. Like when you are wondering about your president's erratic behaviour.

The second scenario has to do with technical decisions, particularly when trying to find the root cause of an issue. In fact, one reason simpler explanations are often favoured in the scientific environment is that they are more easily testable and therefore falsifiable. i.e., it is easier to quickly rule out simple but valid explanations first.

That’s a useful heuristic you want to apply during incident responses and in other forms of war room situations. Go with the simplest explanation, according to available data, while accepting that future data may rule it out entirely in favour of what might appear to be a more complex hypothesis at the moment.

This is ultimately a sensible pragmatic approach, as the number of complex explanations is theoretically infinite, and each one of them might take a long time to be validated.

Or, to say it as Wittgenstein has put it in his Tractatus Logico-Philosophicus:

The procedure of induction consists in accepting as true the simplest law that can be reconciled with our experiences.²

Validating such hypotheses enriches your experience and may inspire you to formulate new ones with greater sophistication and accuracy.

Of course, you can also apply the same principle of simplicity when designing a new technical solution. This boils down to understanding the difference between inherent and accidental complexity. Inherent complexity is, as the word says, intrinsic to the problem at hand. It cannot be simplified away. e.g., a particular piece of business logic might have complex commercial rules. Conversely, accidental complexity has to do with the unnecessary sophistication introduced as part of a solution. i.e., an event-driven massively distributed system can introduce a lot of accidental complexity if you’re just trying to validate a new product idea for a handful of users.

I hope you had enough of an overview of how to use Occam’s razor in your life.

Let us now move on to my favourite, attributed to a seemingly unknown Robert J. Hanlon.³

Hanlon’s Razor

This is my absolute favourite, and that’s because it helped me make sense of numerous situations I’ve encountered on my journey with leading teams. That, and the fact that I vividly remember the moment I first encountered it. It was in a document, or rather a txt file, that a very talented software engineer I respected and had a wonderful relationship with had written and asked me to review. This was around 2014 or 2015 at the latest, and I remember reading the txt file in my terminal with Vim on one of the coolest laptops that ever existed, a ThinkPad x220⁴.

There are just moments like that that mark your existence forever.

Enough with nostalgia, of which I’m sure you couldn’t care less. Let’s see what the razor looks like in its most popular version:

Never attribute to malice that which is adequately explained by stupidity

Since I fall more in the growth mindset camp than in the opposite fixed mindset one, the version I personally use and share with others uses competence instead of stupidity. Believe me, it helps immensely with reception, even in this time and age when communication norms seem to invite confrontation, judgement and hatred.

I can’t tell you how many people I’ve seen light up when I first introduced them to this razor.

Similarly to what I mentioned with regard to Occam’s, we often fall into the trap of suspecting other people’s intentions and agendas. I’ve been there, more often than I’d like to admit. We are somehow attracted to the malicious explanation, as it reinforces our belief that someone we are in conflict with might be mean or have bad intentions. Incompetence is a more common and plausible explanation in many cases. It’s also the simplest explanation, back to Occam.

The good thing about Hanlon’s razor is that once you’ve seen it, you won’t unsee it.

It will help you better understand why someone on your team is often delivering low-quality work. Why your manager is failing to provide you with meaningful feedback. Or why the head of HR is just messing up every single decision and communication.

Believe me, and Hanlon: in most cases that’s just incompetence.

And there is light at the end of the tunnel, because there is little you can do to face malice, besides covering your ass, but there are plenty of proven ways to deal with incompetence.

In most cases you’ll want to leverage one form or another of education, something you can do with people around you regardless of their title relative to yours. Sometimes, you just want to show those who need to act on it the incompetence. Incompetence is a lot easier to prove than malice or intention.

You can’t manipulate your way out of not being able to craft a solid roadmap or even something as simple as calculating the result of 2+2.⁵

As an Italian, there’s one more thing about Hanlon’s razor that I like. It is in stark contrast with a popular statement attributed to Giulio Andreotti.

For those unfamiliar with the person, think about a combination of a Machiavellian politician mixed with strong Catholic beliefs comparable with the most strict cardinals and ministry of the church. He’s no longer alive, but he dominated most of the Italians’ political history in the second half of the twentieth century.

In other words, not in my top ten list for political role models.

And here is what he famously said.

A pensare male si fa peccato, ma spesso ci si azzecca.

Loosely translatable as “It’s a sin to think ill of other people’s intentions, but it’s usually correct/spot on”.⁶

I’ll save you all the interpretations about the role of sin in the Catholic religion and how it’s been used to control people. What I’m more interested in is that Andreotti’s view is one that suggests we’re supposed to judge people’s intentions (which is objectively a sin, and as they say, it’s a prerogative of God). Not only that, but it’s somehow an apology for mischievous behaviour. If everyone is correct in assuming everyone else's bad intentions, that includes me (or Andreotti in this case). If he believes in his statement, he's implicitly admitting his guilt too.

Quite the vicious circle, if you believe me. Who would want to live in a world like that?

I much prefer the agency of taking incompetence as an explanation. Or, more broadly, ignorance.

Something for which many solutions exist that don’t require people mistrusting each other.

---

WIT Promo for Q1 2026

I’ve recently decided to resume offering quarterly promos for people who are willing to benefit from my services.

I’m happy to announce that I’ve opened up the Q1 promo that will run until the end of March 2026.

I’m making it easier for Women In Tech to level up their engineering leadership skills by offering an exclusive discount to the Sudo Make Me a CTO: 30% off for the first 12 months.

You can find out all the details at the official promo page, or by clicking the button below.

Feel free to share this opportunity with people you know, and do not hesitate to reach out if you’d like to learn more about it.

You can always schedule a free 30-minute session to get all your questions addressed.

Looking forward to seeing the community grow with more diversity.

1. It seems like in the past couple of years everyone has learnt this new word. Therefore I’m assuming it’s familiar enough for you all to understand that a) it means making up things that aren’t necessarily true or verified and b) I’m being sarcastic

2. This is the kind of cool thing you can learn if you read the entirety of a dense Wikipedia page. Like the fact that Elliot Sober stated that philosophers may have made the error of hypostatizing simplicity. Can you even pronounce that without stuttering?

3. It's surprising to notice how much shorter the Wikipedia page dedicated to Hanlon's razor is compared to the one for Occam. But the fact that it's mentioned in the Jargon File, aka The Hacker's Dictionary, largely compensates for the difference.

4. I got that laptop in 2012, and I still have it today. It sits in the garage for the occasional web search or YouTube video when I’m engaged in DIY activities.

5. OK, I know this can be questioned in at least two ways. The first one is the Orwellian approach of repeating fake things until they become true, the famous 2+2=5. Sorry, but I don't have a solution to offer against fake knows except for critical thinking. The second, fancier and a lot more hyped, is that people can delegate some of that work to statistical models that are good at generating plausible text. The problem, though, is that they will not be able to validate that the output is correct. Errors will slip through, and sooner or later incompetence will become evident.

6. Thanks to the folks here who served as an inspiration and input for the translation.