One of the more recent decisions I have made was whether to replace my current car with a newer SUV. On the surface, it looked like exactly the kind of decision I am good at: a defined problem, a finite set of options, and plenty of data available to compare them. I approached it the way I approach most things, with research, comparisons and a set of criteria written down before I started looking seriously.

What I did not expect was how quickly that structure would start working against me, or how long it would take me to notice.

How It Started: The Rational Model

In the beginning, the process was genuinely rational. I identified the problem, established what mattered to me, and started gathering information on each alternative. The criteria were the obvious ones, and I could have defended every single item on the list to anyone who asked.

• Reliability — how likely it is to spend time in a workshop
• Safety — ratings, systems, and how it performs in a collision
• Purchase price — what it actually costs to get into the car
• Financing — the terms, the rate, and what the monthly figure does to everything else
• Resale value — what it will be worth when I want to move on from it
• Long-term ownership cost — servicing, parts, insurance, fuel
• Fit for purpose — whether it genuinely suits how I use a car

This is the rational model of decision-making almost exactly as it is described: define the problem, establish the criteria, gather the information, compare the alternatives, then select the option that scores best. It is a clean process, and on paper it should have produced a clean answer.

Where Logic Quietly Gave Way

The more time I spent researching, the more I noticed something the criteria list had no column for. I was thinking about how enjoyable each car would be to drive. How modern the interior felt. The technology, the performance, the way the thing looked in a car park. And, if I am being honest about it, how I would feel owning it.

At one point I was comparing sensible, well-evidenced options like Toyota and Kia against newer alternatives like Geely. Two entirely different questions were running in my head at the same time, and they were not asking for the same answer.

“

One part of me was asking which car would be the most reliable and financially sensible over several years. Another part was asking which one I would actually enjoy owning. Both of them were me.

That is the part worth sitting with. It was not that emotion crept in and corrupted an otherwise clean process. The emotional criteria were there from the start; they simply were not written on the list, so they never had to justify themselves the way reliability and resale value did. Anything that avoids scrutiny tends to carry more weight than it has earned.

Why the Rational Model Is Harder Than It Looks

The rational model assumes a few things that simply do not survive contact with a real decision. It assumes you can obtain enough accurate information, evaluate all the available alternatives, and reasonably predict the consequences of each one. Every one of those assumptions broke down on a decision as ordinary as buying a car.

None of this is a failure of effort. It is the nature of the problem. No amount of additional research was going to tell me what a particular model would be worth in five years, or whether the specific car I bought would be the reliable one or the unlucky one.

Bounded Rationality Describes It Better

Bounded rationality is a far more honest description of what I was actually doing. It recognises that people make decisions using simplified versions of complex problems, because our information, our time and our capacity to process everything are all limited. We do not solve the real problem. We solve a manageable version of it and hope the simplification was a fair one.

In practice, that meant I never evaluated the market. I reduced it. I set a handful of non-negotiable requirements, used those to cut the field down to a small group of candidates, and then compared that group properly. The comparison inside that shortlist was rigorous. The decision about what made the shortlist in the first place was much rougher, and it is the step nobody writes down.

That is worth being aware of, because the shortlist is where most of the decision actually gets made. By the time you are carefully weighing three cars against each other, you have already discarded everything else, often on instinct and a few minutes of reading.

Maximising Versus Satisficing

The other thing I recognised, later than I would like to admit, was that I had been trying to maximise. I wanted the car that was best on reliability and best on price and best on features and best on driving experience and best on safety and best on resale value, all at once.

The problem with that goal is not that it is ambitious. It is that it is incoherent. No such car exists. Those criteria trade against each other by design, and a vehicle that led on every one of them would not be a bargain, it would be a contradiction. I was searching for something that could not be found, which is an excellent way to keep researching indefinitely and never decide anything.

Satisficing is the more realistic approach: deciding in advance what an acceptable standard looks like, then choosing the option that meets it well enough rather than continuing the search for a perfect result that does not exist. It sounds like settling. It is not. It is recognising that past a certain point, more searching stops improving the decision and starts merely delaying it.

Emotion Was Never the Real Problem

Looking back, I do not think the emotional side of this decision made it a bad decision. A car is something I would potentially use every day for years. Enjoyment is not a frivolous criterion in that context; it is a real part of the value I get from the purchase. Pretending otherwise would have produced a more defensible spreadsheet and a worse outcome.

The actual mistake would have been letting enjoyment influence the choice while telling myself, and anyone who asked, that the decision was purely financial. That is where an emotional decision becomes a dishonest one. Not because emotion was present, but because it was disguised as logic, and therefore never had to compete fairly against the things that mattered more.

“

Being rational does not mean removing emotion from a decision. It means knowing where the emotion is sitting, and making sure it is not quietly outranking the things that matter more.

Ali Abdisalam Hussein

What I Would Do Differently

The specific failure in my process was one of sequencing. I started looking at cars before I had finished deciding what I wanted from one, which meant my criteria were being quietly rewritten by whatever I had just seen. A car with an impressive interior did not simply score well on my list; it made interior quality more important on the list.

Setting the priorities before becoming attached to a particular option makes it far easier to tell the difference between what I genuinely need and what I simply want. Both can still count. They just need to be counted separately.

My Main Takeaway

The bigger lesson is that I make far fewer completely rational decisions than I believe I do. I use spreadsheets, comparisons and research, and all of that is real work that improves the outcome. But personal preference is still sitting somewhere in the final choice, and the structure around the decision can make that easier to miss rather than easier to see.

Understanding bounded rationality has made me more comfortable with that, not less. The goal was never to make the perfect decision, because the information required to identify a perfect decision is not available to anyone. The goal is to make a well-informed and defensible decision with what I actually knew at the time, and to be honest with myself about which parts of it were logic and which parts were preference.

That standard is achievable. Perfection was never on the table, and chasing it mostly just costs you the time you could have spent deciding.