Two sentences that look the same on paper
Here are two lines that could sit in the same section of the same résumé.
The first: owned vendor selection for a forty person engineering organization.
The second: kept a manual process running for two quarters past the mandate to automate it, because every failure in the automated version was landing on the same two people, and those two people were the ones who would have to fix it at midnight.
The first line describes territory. It names an area that was yours, a box on a chart with your name in it, a thing you were accountable for over a span of months. It is true, it is verifiable, and every other person who has ever held that job could write it without changing a word.
The second line describes a moment. Something was open, you closed it, and it could have closed the other way. There was a reason, the reason was yours, and somebody senior to you may well have disagreed.
One of these is a responsibility. The other is a decision. They look similar enough on a page that most people treat them as the same kind of evidence, and the confusion is expensive, because almost every instrument in hiring collects the first kind while asking for the second.
A responsibility is a standing condition. It persists. It describes what was true about your role for as long as you held it, and it can be inferred from your title by anybody who knows the field. A decision is an event with a before and an after. It has a date, it has an alternative that did not happen, and it exists nowhere except in the memory of the person who made it.
What your own memory does to the evidence
The first obstacle to writing down a decision is not modesty and it is not confidentiality. The obstacle is that your memory has already done something to it.
Baron and Hershey ran five studies in which people read descriptions of decisions made by others under uncertainty, some medical and some monetary, and then rated the quality of the thinking behind them, the competence of the person deciding, and their own willingness to let that person decide on their behalf. Everyone was told plainly that they had all the information the decision maker had. The ratings still moved with the result. Thinking was rated as better, and the decision maker as more competent, when the outcome happened to be favorable. In the monetary gambles the effect went further still: the thinking was rated as better when the option that had not been chosen turned out poorly. Asked directly, the raters said that outcomes should not count in an evaluation of this kind. They counted anyway (Baron & Hershey, 1988).
Now hold that beside the older finding underneath it. Fischhoff showed that simply knowing how something turned out raises the likelihood people assign to it in retrospect, changes which facts they treat as having been relevant, and does both regardless of how likely the outcome was. The part that matters most for a working life is what came next. His participants were largely unaware that outcome knowledge had done anything to their perception at all, and so they overestimated what they would have known without it, and overestimated what other people had known without it. He argued that this lack of awareness restricts the ability to judge or to learn from the past (Fischhoff, 1975).
You are the historical judge of your own career, and you know how all of it turned out. Every call you made has been rewritten in your memory by what happened afterward. The ones that worked now feel as though they were obvious at the time, which makes them feel unimpressive and not worth mentioning. The ones that failed now feel as though you should have seen it coming, which makes them feel disqualifying. Both of those feelings are the bias doing its work, and both push you toward the safest thing on the page, which is the responsibility.
Why you cannot narrate a process you never ran
There is a second obstacle, and it is stranger than the first, because it hits experienced people hardest.
Klein and his colleagues studied fire ground commanders making decisions under time pressure where lives and property were at stake. They interviewed twenty six commanders with an average of twenty three years of experience each, working through nonroutine incidents that demanded real expertise, and examined 156 decision points in the process. In fewer than twelve percent of those decision points was there any evidence of two or more options being compared side by side. In over eighty percent of them, the commander recognized the situation as typical of a familiar pattern and recognized a course of action as typical for that pattern, and moved. The model the researchers built from this became known as recognition primed decision making, and it puts recognition rather than calculation at the center of expert choice (Klein et al., 2010).
Sit with what that means for the interview question you have been asked in some form at every stage of your working life. When somebody asks you to walk them through how you weighed the options, the truthful answer for an experienced person is very often that the options were never laid out to be weighed. You saw what kind of situation it was, and the shape of the response came with the recognition.
That answer is hard to say out loud, because it sounds like an admission of having no process. So people reach for something that sounds more like a process, and what is closest to hand is the responsibility: the area you owned, the scope you had, the size of the budget you were trusted with. It is safe, it is describable, it is verifiable, and it says nothing about you that the job title did not already say.
The private half
I want to give this a name, because once it has one you will see it in every career document you touch.
Every decision you have ever made has a public half and a private half.
The public half is the part the systems kept, which is the outcome, the date, the project name, the headcount, the number that moved, the line in the performance review and the slide in the quarterly deck. It is recorded somewhere, it is mostly verifiable, and it is the raw material of every résumé anybody has ever written.
The private half is everything that was true inside your head at the moment of choosing. It holds what you knew, and what you did not know and could not find out in the time you had. It holds what you could have waited for and chose not to wait for. It holds the consideration that mattered most to you while nobody else in the room was weighting it the same way, and it holds the reason you moved.
The private half was never written down. Not hidden, not withheld, and not protected by any agreement you signed. There was simply no system collecting it, because no organization has ever built one, and the moment passed while everybody involved got on with the next thing. It exists in exactly one place.
That is what makes it worth something now. The public half of your record is reproducible by anybody who held a similar role. The private half is not reproducible by anyone at all.
When your own judgment counts as evidence
There is a failure mode waiting at the end of this argument, and naming it is what keeps the rest of it honest.
If the private half is the valuable part, and the private half is only in your head, then nothing stops a person from simply asserting good judgment. Everybody believes they have it. The belief is not worth anything on its own.
Kahneman and Klein came at this problem from two traditions that had been expected to disagree, one built on heuristics and biases and the other on naturalistic decision making, and they ended up agreeing on the boundary. Evaluating the likely quality of an intuitive judgment, they concluded, requires two assessments: how predictable the environment is in which the judgment was made, and how much opportunity the individual had to learn the regularities of that environment. Their sharpest line is the one to keep. Subjective experience is not a reliable indicator of the accuracy of a judgment (Kahneman & Klein, 2009).
This converts directly into a test you can run on your own sentences. The claim that you simply knew is worth nothing, because the feeling of knowing is not evidence and the research says so plainly. The claim that you had seen this specific failure three times in four years, and recognized it in the first fortnight the fourth time, is worth a great deal, because it names the regularity and it names where the learning came from. One sentence asserts a gift. The other describes an environment that taught you something and a person who was paying attention while it did.
That is also the difference between a decision story that a skeptical reader can use and one that reads as self flattery. You are not asking to be believed about your instincts. You are showing the conditions under which those instincts were trained.
The test, and the three things a decision sentence has to carry
The interview already asks for this, in every version of the question about a time you had to decide without all the facts, and the question is rarely what goes wrong. What goes wrong is that people arrive at it with nothing prepared and improvise, and improvising under outcome bias produces the polished version rather than the true one.
So prepare it. A decision sentence has to carry three things, and if any one of them is missing the sentence collapses back into a responsibility.
What you knew. The specific state of information at the moment, which is almost always partial. Half the information is the normal working condition and not a confession of anything.
What you did not know, and could not get. This is the part people cut, because it feels like weakness. It is the opposite. It is what makes the choice a choice rather than an arithmetic problem, and it is the single clearest signal that you are describing something real rather than reconstructing something flattering.
Why you moved anyway. This is the reason as it stood before the outcome existed, rather than the outcome working backward to supply one.
Then apply the filter that strips the bias out, which is to ask whether you would make the same call again on the information you had at the time. The question is not whether it worked. The question is whether you would make it again.
That filter does something useful and slightly counterintuitive. It admits decisions that failed. A call you would still make on the same information, that went badly anyway, is frequently stronger material than a call that succeeded, because a person who can hold those two facts together at once is demonstrating the exact thing every hiring manager is trying to find out and cannot ask about directly. It also excludes a whole category of things that feel like achievements and are not, which is the decisions that worked out for reasons you did not foresee and have quietly taken credit for ever since.
Objections worth taking seriously
My decisions were all small. Nearly all decisions are small, including the ones that turn out to matter most. Scale is a property of the organization, not of the judgment. The call about which two people would absorb the failures of an automated process is small by any measure a chart would recognize, and it is better evidence about how you think than a budget number ten times its size, because the budget number was handed to you by your level and the call was not.
I cannot discuss the specifics. Very little of the private half is confidential once you look at it closely. The confidential parts are usually the client, the numbers, the system names, and the people. The structure of the reasoning is almost never the protected part. You can describe what you knew, what you could not find out, and why you moved, without naming any of the four, and the sentence loses nothing that mattered to a reader.
Would you make it again sounds like a way of excusing bad outcomes. It would be, if it were the only test. It is not a test of whether the decision was good. It is a test of whether you are evaluating the decision or the result, which is the thing Baron and Hershey showed people cannot do reliably even when they are told to and want to. Apply it and you will find it disqualifies plenty of your own calls, including some you have been proud of for years. A filter that never rejects anything is not a filter.
Nobody asks me about decisions anyway. They ask constantly, in the form of questions about a difficult situation, a disagreement, a time something went wrong. Those are requests for the private half, phrased as requests for a story. What they get back, most of the time, is a responsibility with a narrative arc attached.
Where the material comes from
The private half of your record is not lost. It is unwritten, which is a different condition and a recoverable one.
Take one job you have held and one call you made inside it where the information ran out before the decision was due. Write down what you knew at that moment, in a sentence. Write down what you could not find out in the time available, in a sentence. Write down why you moved, in a sentence, and hold yourself to the reason as it existed before you knew the outcome. Then read the three sentences back and ask whether you would make the call again on the information in the first two.
Most of what you write this way will never appear in a document you send to anybody. That is not a waste, and it is not the point. The point is that the reasoning becomes available to you again, in words, on demand, instead of sitting compressed inside a result you have been carrying around for years. A career spent deciding under uncertainty leaves almost no trace of the deciding. Every system you worked inside recorded the outcome and discarded the reasoning, and the reasoning was the part that was yours.