Our cognition is limited. Over years of evolution and adaptation our brains have learned to create shortcuts for decision-making.
I’ve made bad decisions because of shortcuts I’ve unconsciously taken:
I hired people because I liked them despite some gaps in their skills
I waited for too long to sell investments resulting in a loss
I kept people on for too long because of how much time I had spent on them
and many more...
These shortcuts helped us extremely well in making quick decisions and surviving on the Savannas. But we didn’t evolve to handle complex decision-making, systems thinking, or long-term planning. That’s why we have these built-in cognitive biases.
This is just a starting point — a set of cognitive biases is a deep field with entire books written on them. But this overview can serve as a helpful reminder.
Here are the most impactful ones while making important decisions, supported by examples from engineering leadership:
1. Confirmation Bias
We like to be right, so we seek confirmation of our beliefs. Anything that can confirm our beliefs is used to support them.
I was using Ruby on Rails for web development for years, so while deciding on technology, I looked for any possible confirmations that it's the best choice out there, often ignoring that it might not be the best solution.
Overcoming:
Prepare clear decision criteria and make sure your personal experience does not blur it.
Play Devil's advocate and challenge yourself with the question: "What would change your mind?"
Doing a pre-mortem exercise when we assume that a project has failed before even starting it, and list what went wrong, maybe your favourite framework wasn't ideal?
2. Anchoring Bias
We tend to pay more attention to the first thing we've heard of.
When an experienced senior engineer says during the meeting that something is easy and will take 2 days to finish. It becomes a reference point for the rest of the group, that's our "anchor".
Overcoming:
When discussing estimations in a group, making hiring decisions, or handling any situation where I wanted to avoid anchoring to a reference point, I asked everyone involved to write down their opinions first, then share them simultaneously in the channel before we started the discussion.
3. Hindsight Bias
It's so much easier to point out what we should have done in the past, knowing the results. But when we are in the moment, with the same data, we would probably act the same as we did.
I’ve heard this too many times: “The company did layoffs, obviously, they should have done more cuts a year ago, etc.” — It's so easy to judge knowing the future.
Overcoming:
When making decisions, list everything that you know at this stage just before making the call. The truth is that given the data from the past and the same context, chances are I'd do exactly the same.
4. Recency Bias
The recent events are fresh in our memory, so we give them too much weight.
While doing performance reviews of your team members, you might put much more emphasis on their recent performance, just a few recent weeks, than the full scope of the review, like 6 months or more.
Someone who had an amazing, productive last month might be perceived better than someone who was the key player, but a few months ago.
Overcoming:
Document what is changing and happening over the given timespan, and make snapshots. Do not focus only on recent events.
For general perception, it's important to remember that if something happened recently, it will have higher weight, as it's just easier to recall by our brains.
5. Groupthink
We want harmony. We want people to like us. Saying something that breaks that harmony might not be perceived well (especially in Japan), but maybe it should be! Groups tend to silence controversial voices.
Overcoming:
Rewarding an opposite stance serves as a great reminder for others that it's good to have a different opinion. It brings more perspectives to the table.
On the AAR sessions, or post-mortems, it's important to work separately and reveal your ideas and feedback altogether. If we can’t see beforehand what others think, groupthink can be tamed
6. Sunk Costs Fallacy
We are stuck in positions of investments, jobs, and relationships because we have invested our time, money or effort.
This is a strong force that keeps us in for years, and the longer we postpone resolving it, the worse it gets.
Overcoming:
Ask yourself if you would pick right now to join this company, would you still join? Replace "company" with any other aspect that you have invested in.
I like this Chinese proverb, although it’s not about stopping but starting:
The best time to plant a tree was twenty years ago.
The second best is now.
7. Overconfidence Bias
Studies show that 80% of drivers think they're above the average. This is quite an overconfidence, and it happens in different domains too. Entrepreneurs believe that they will succeed at a high rate, even if the chances are relatively low.
A project you led in the past may have gone well, but new projects can bring different problems and unknown things you don’t yet know about.
Overcoming:
Take your past as an example, how many times were your decisions flawless and without any mistakes? Be honest with yourself — it rarely happens.
Remember that it's quite probable that you are among average drivers, and the same goes for other domains.
Being humble in understanding and learning from your mistakes can help with overcoming this bias.
Being highly motivated is not the same as being overconfident. In our society, we often confuse confidence with real skill, and this is important to distinguish one from the other.
Closing Thoughts
I believe that when you know these biases and start to notice them, you can make better choices. This is one of the most important tools in your decision-making toolbox.
Which of these biases do you think is the most impactful?
Thanks for reading!
— Michał
Post Notes
Discover Weekly — Shoutouts
Articles that might help you explore new perspectives, which I’ve read recently:
How Social Media Shortens Your Life by (it’s bold!)
Defining Your Core Values (The Compass) by
The State of Software Development in 2025 by
Connect
LinkedIn | Substack DM | Mentoring | X | Bsky
Great summary – the sunk cost part especially hit home. Sometimes the smartest move is to just walk away.
Would love to hear if any of these biases had a real impact on your team or a project you led. Any story that stuck with you?