Day 1 · Critical Thinking: How to Be Right More Often
Why capable, experienced people get confidently wrong
You cannot think your way out of a bias you do not believe you have. That is what makes bias the hardest problem in this whole series, and the right place to begin.
We tend to imagine that clear thinking is something you either have or you do not, and that careful, intelligent people mostly have it. The uncomfortable truth is that intelligence offers almost no protection. The mind runs a set of fast, automatic shortcuts that were useful for surviving on a savannah and are quietly unreliable in a requirements workshop. These shortcuts do not feel like shortcuts. They feel like seeing clearly. And the more experienced you are, the more confidently they fire, because your pattern recognition is faster and your certainty arrives sooner. Daniel Kahneman spent a career studying this and described two systems in the mind. One is fast, intuitive and effortless, and it produces most of your snap judgements. The other is slow, deliberate and lazy, and it only engages when you make it. Nearly every bias in this piece is the fast system answering a hard question with an easy one while the slow system dozes. You cannot switch the fast system off. But you can learn to notice when a decision deserves the slow one, and that noticing is most of the skill.
The four biases that cost analysts the most
There are dozens of documented biases, but four do the most damage in our kind of work, and they are worth knowing by name, because a bias you can name is a bias you can catch.
The first is confirmation bias, the pull to notice and believe the evidence that supports what you already think, and to quietly explain away the rest. A team decides customers are leaving because of price, and from that moment, every price complaint is remembered, every other reason is filed under noise, and the theory grows stronger not because it is true but because nobody was looking for the parts that would break it.
The second is anchoring. The first number said out loud in a room silently drags every later number toward it. Someone guesses the work is about six weeks, and now every estimate in the session orbits six weeks, including yours, even though the guess was made before anyone understood the work.
The third is the sunk cost fallacy, the reason so many doomed projects limp on. Because a fortune has already been spent, stopping feels like admitting waste, so more is spent to protect the first amount, and the past, which cannot be recovered, holds the future hostage.
The fourth is the availability heuristic, judging how likely something is by how easily an example comes to mind. One dramatic incident last quarter dominates the risk register for a year, while a quieter, more probable risk goes unlisted because nothing memorable ever forced it into view.
Why expertise makes it worse, not better
It is worth sitting with how invisible these are from the inside. I once watched a planning session where the first person to speak floated a rough guess of two weeks for a piece of work nobody had scoped yet. For the next hour, every estimate offered landed somewhere between ten days and three weeks. When the work was finally broken down properly, it was closer to two months. Nobody in that room felt anchored. They felt like they were reasoning. The two-week guess had simply become the gravity the whole conversation fell toward, and the only thing that would have broken its pull was someone noticing it was there and saying so out loud before the numbers started.
It would be comforting to think these biases fade with experience. They do not, and there is a specific reason. Expertise is built on pattern recognition, on having seen enough that the answer arrives before you have consciously worked it out. That is genuinely valuable, and it is also exactly the machinery that produces confident error. The expert's fast system offers an answer so quickly and so fluently that the slow system sees no reason to check it. A newcomer, unsure, is sometimes forced into the slow, careful thinking that the expert has learned to skip. This is why the phrase to fear most in your own head is "it is obvious". Obvious is the feeling of the fast system refusing to be checked.
Why trying harder does not work
The natural response, once you accept you are biased, is to resolve to be more objective, to try harder to see straight. It does not work, and understanding why saves years of wasted effort. Bias is not a lapse in concentration. It is the normal operation of a fast, efficient mind, and you cannot out-concentrate it any more than you can out-concentrate an optical illusion. Staring harder at the two lines does not stop the one that looks longer from looking longer. What changes the answer is not effort but a different process: a ruler laid against the lines, a step that checks the intuition from the outside. The same is true in your work. You do not beat your biases by being determined to be fair. You beat them by building small habits that check your thinking, whether or not you feel biased at the time, because the whole trouble is that you never feel biased at the time.
How to catch yourself
You cannot delete your biases, but you can build habits that interrupt them at the moment they matter. These are small, deliberate acts of slowing down, aimed at your own thinking rather than anyone else's.
- Ask what would change your mind, and whether you have actually looked for it. If the honest answer is nothing and no, you are not holding a conclusion; you are defending a belief.
- Argue the opposite case out loud before you decide. Make yourself state the strongest version of the view you are about to reject. If you cannot, you do not understand your own position well enough to trust it.
- Separate the feeling of certainty from the evidence for it. Write down what you actually know versus what you have assumed. Certainty is an emotion, not a measurement.
- Get one fact on the table before the first opinion anchors the room. A single real number, spoken early, protects a whole meeting from orbiting someone's guess.
- Name the bias out loud when you suspect it. Saying this might be sunk cost talking does more to break its grip than any private resolve, because it gives the whole room permission to check the thought too.
The mistakes that keep you certain and wrong
The failures here are quiet, because they feel like competence while they happen.
- Believing bias is other people's problem. The moment you feel immune is the moment you are least protected, because you have stopped watching yourself.
- Treating confidence as a signal of accuracy. How sure you feel and how right you are are only loosely related, and experience widens the gap.
- Gathering evidence only after you have already decided, then calling the pile of it proof.
- Letting the loudest or earliest voice set the anchor, and mistaking the resulting agreement for a conclusion the group actually reasoned its way to.
Why this is your real edge
Nobody notices their own anchoring on a CV, and no job description asks for it. But it is most of what separates the analyst who gets handed the ambiguous, high-stakes, politically loaded problem from the one who gets the tidy, well-scoped one. The first is trusted because, over the years, they have been wrong less often in the moments that mattered, and they were wrong less often because they learned to distrust their own certainty before it hardened into a plan. That habit never shows up as a dramatic save. It shows up as a career where fewer things blew up than should have, and people are slowly deciding that your judgement can be relied on when the stakes are real. That reputation is the most valuable thing you will build, and it starts with the willingness to doubt yourself on purpose. Tomorrow, the frame, and how the way you state a problem quietly decides which answers you are even able to see.
Go out and be successful.
Oluwatosin Ogunkoya | Flotog BA Insights | www.flotogbainsights.com
Tomorrow: The Frame Decides the Answer - How the way you state a problem controls which solutions you are able to see at all.