Why this subject is usually worthless
It is worth conceding the case against this whole subject before making any argument inside it, because the case is strong and most writing about decisions pretends it does not exist.
Ask anybody to name the decisions that made their life and they will produce a list. The list will be short, it will be coherent, and it will lead in a reasonably straight line to where the person is sitting. That coherence is the problem. The list was not found in the past. It was assembled in the present, by the person those decisions produced, working backward from a known destination, and a list built that way will always look like a life that was steered.
This is why the genre is full of nonsense. A story assembled from the end has one enormous advantage over the life it describes, which is that it knows how things turned out, and it will therefore promote every event that pointed here and quietly drop every event that pointed elsewhere.
So the useful question is not what your decisions were. It is whether any filter exists that a backward-assembled list cannot pass. There are four, they run in sequence, and each one takes a substantial cut out of the list you started with. What is left at the end is small, and it is the only part worth anything.
The first cut: the ones that were never decided
Start by removing everything that was a default.
The cleanest demonstration of what a default does comes from organ donation. Johnson and Goldstein set the effective consent rates of European countries beside the rule each country uses. In countries where a citizen is a donor unless they say otherwise, consent runs at levels that countries requiring citizens to register cannot approach, and the gap between the two groups is far larger than any plausible difference in what those populations believe about donation (Johnson & Goldstein, 2003). Nobody in either group experiences themselves as having been shaped by a form. The rule did the deciding, and the citizen supplied the feeling of having a view.
Your life is full of this. The town you stayed in because leaving would have required an act and staying required none. The field you continued in because the next step was already drawn. The arrangement at home that was never negotiated and simply became the arrangement. The savings you do or do not have because of what an employer set up on your behalf in your twenties. Every one of those looks like a decision from the inside, since you can produce a reason for it on request, and the reason will be sincere and will have been constructed afterward.
There is a second process worth separating out here, which is habit rather than default. Ouellette and Wood distinguished two routes by which past behaviour shapes future behaviour. Where a behaviour is well practised and the context stays constant, the processing that initiates it becomes automatic, and past frequency reflects habit strength and directly drives future performance. Where a behaviour is not well learned, or the context is unstable or difficult, conscious decision making is required, and past behaviour feeds intentions rather than acting directly (Ouellette & Wood, 1998).
That distinction is doing real work for this article. It says that the regime you are in depends on how practised the act is and how stable the setting is, which means the events in a life that were truly decided are the rare ones: unpractised, in an unstable setting, where nothing automatic was available to carry you. Everything else was running on rails, and the rails were not yours.
The second cut: the ones with no live alternative
Of what survives the first cut, most fails the second, and this is the test the rest of the article is built on.
The live alternative is the other thing you could have done, which you can still name, and which you can still say something specific about. Not a theoretical alternative. A particular one, available at the time, with its own consequences that you had thought about. Where there is no live alternative, there was no decision, whatever it felt like.
The test is easier to run than it sounds, because a live alternative leaves a residue. You can say what it would have cost. You can name who was arguing for it. You can describe, without much effort, the version of your life in which you took it, and the description will have detail in it rather than atmosphere. Where you reach for the alternative and find only a vague other path, you are looking at something that was accepted rather than chosen.
This cut removes a category people are unusually attached to, which is the forced move. The redundancy, the illness, the family obligation that arrived without being asked for, the relocation that came with somebody else's job. Those are enormous events and they are formative, and they are not decisions, because the alternative was not live. What you did afterward may well have been. The event itself was weather.
It also removes most of what people describe as the decision to keep going. Continuing is the default condition, and it becomes a decision only at the specific moments when leaving was live, costed and on the table.
The third cut: the ones you would not make as the person you are now
Now the filter that separates this from every other piece of advice about examining your choices.
The obvious move here is to ask whether you would make the decision again on the information you had at the time. That is a good test and it does a specific job, which is to stop you grading a decision by its result. It is not the test this article is after, because a life is not assessed the way a professional call is.
Ask instead whether you would make it again as the person you are now.
The reason that question carries information is a finding most people get backwards about themselves. Quoidbach, Gilbert and Wilson measured the personalities, values and preferences of more than nineteen thousand people between eighteen and sixty eight, and asked them either to report how much they had changed in the past decade or to predict how much they would change in the next one. The design is what makes it convincing: for each decade of life they compared what people of a given age predicted about their next ten years against what people ten years older reported about the ten years just gone. Young, middle aged and older people all said they had changed a great deal in the past and would change relatively little from here. The authors put it plainly, that people regard the present as a watershed at which they have finally become the person they will be for the rest of their lives (Quoidbach et al., 2013).
Their illustration of the cost is hard to improve on. Young adults pay to remove the tattoos that teenagers paid to get, middle aged adults hurry to divorce the people whom young adults hurried to marry, and older adults visit health spas to lose what middle aged adults visited restaurants to gain (Quoidbach et al., 2013).
Hold that beside the repeat test and the mechanism becomes clear. The person who made the decision and the person evaluating it now are measurably different people, whether or not either one believes it. A decision that both of them endorse has therefore survived a real test, because it has been ratified twice by two different sets of values. A decision that only the earlier one endorses has not failed morally and is not something to be ashamed of. It has simply told you that it belonged to that person and to that period, which is useful information and a different kind of thing from evidence about you.
Why the ones that worked are the weakest of all
There is one more cut, and it takes out precisely the decisions people most want to talk about.
Iyengar, Wells and Schwartz followed students through a real job search in their final year, having first measured how far each one tended to maximise, meaning to search for the best available option rather than a good enough one. The maximisers did better on the outcome by an unambiguous margin, securing jobs with twenty percent higher starting salaries than the students who satisficed. They were also less satisfied with the jobs they got and carried more negative affect through the whole search, and those effects were mediated by their greater reliance on external sources of information and their fixation on the options they had taken and the ones they had not (Iyengar et al., 2006).
That study is about job hunting, and the finding that carries beyond it is structural rather than topical. Outcome quality and the quality of a decision as an expression of the person deciding are separable, and they can move in opposite directions in the same choice. A decision can be the best available on every measurable axis and still be one that the person who made it does not recognise as theirs, because the reasons driving it came from outside.
This is why the decisions that worked are the weakest material in your record. They are compromised as evidence by their own success. You have had years to construct a reason for each one, the reason has never been tested against a contrary result, and the outcome supplies so much confirmation that you can no longer see whether the reason came from you or from the room you were standing in.
What survives, and what it is evidence of
What passes all four cuts is a short list, and most people, working properly, find between three and six items on it across an entire adult life. That number will feel too small. It is worth sitting with rather than arguing against, because the smallness is the finding.
What the survivors are evidence of is narrower than a life story and more useful. They are the occasions when your own reasons were the operative ones.
Sheldon and Elliot's self-concordance model gives that phrase a technical meaning worth borrowing. Self-concordant goals are those consistent with a person's developing interests and core values, and the model assigns them a dual role. People pursuing self-concordant goals put more sustained effort into them and are therefore more likely to attain them, and those who attain them get greater well-being from the attainment, mediated by daily experiences of autonomy, competence and relatedness accumulating over the period of striving. The model fitted three longitudinal data sets and held independently of self-efficacy, implementation intentions, avoidance framing and life skills (Sheldon & Elliot, 1999).
Read the survivors of the four cuts as the closest thing you have to a record of self-concordance. Each one was unpractised and unstable enough to require deciding, had a live alternative you gave up, and is still endorsed by a later person with different values. Whatever was doing the work in those moments has now been observed operating across decades and across two versions of you.
That is the evidence hiring never sees, and the reason it never sees it is structural rather than unfair. Selection collects what organisations record, which is roles, outcomes and dates. Nothing in a life is collected. The consequence is not that you should put your divorce on a résumé. It is that the most reliable information about how you decide under real stakes exists only in a place nobody has ever asked you to look, and that the asking has to be done by you.
Four places this is wrong, or nearly
The four cuts would empty the list of anything that matters. They will certainly empty it of most of what you would have said first, which is the intended effect of a filter. The events removed are not being denied. A forced move that reshaped a life is still the most important thing that happened to you; it is simply not information about how you choose, and the article is about one narrow thing.
Would make it again is a way of forgiving yourself. It would be, if it were applied on its own. Coming fourth in a sequence, after defaults, live alternatives and the difference between two selves have each taken their cut, it is doing the opposite work. Most people find it disqualifies decisions they are proud of and preserves one or two they have been uneasy about for years.
Nobody can say what they would do as a different person. You are not being asked to simulate anybody. You are being asked what you think now, with the values you hold now, about a choice made under conditions you can describe. The end of history finding says you will underestimate the distance between the two of you, and underestimating a gap is not the same as being unable to look across it.
My most defining decision is one I regret. Then it did not pass the third cut, and that is worth something on its own terms. A decision you would not make again, examined properly, tells you which of your values changed and roughly when, and that is one of the few ways anybody ever finds out.
The first one
There is one decision in almost every life that comes pre-filtered, and it is the obvious place to start.
The first big decision you made entirely alone passes the first cut automatically, because there was no default available: whatever the arrangement had been until then, it had just stopped applying. It usually passes the second, because the alternative was not theoretical, and in most cases somebody was actively arguing for it. It is far enough back that the person who made it is unmistakably not the person reading this, which means the third cut will give a real answer rather than a comfortable one. And it is old enough that whatever it produced has long since stopped feeling like proof of anything, which is what makes the fourth cut survivable.
It is also the only decision in your life where you can be certain nobody else was carrying any of the weight. Everything after it had somebody in the room, or a precedent, or a rule. That one had you, an alternative, and no way to find out in advance.