In the spring of 1961, a few months into his presidency, John F. Kennedy authorized the invasion of Cuba's Bay of Pigs. By any honest accounting, it was a catastrophe — a CIA-trained brigade of Cuban exiles were routed within seventy-two hours and a new administration's credibility on national security was basically shredded before it had really even begun.
Inside the White House, advisors braced themselves for Kennedy's approval rating to crater.
Instead, eight days later, Kennedy walked to a podium in the State Department auditorium and said, in front of the nation, "There's an old saying that victory has a hundred fathers and defeat is an orphan… I'm the responsible officer of this government, and that is quite obvious." He didn't blame staff or deflect responsibility onto his advisors in the intelligence-community. Instead, he simply owned the loss publicly.
Not only did his approval rating not crater, it surged. Gallup's late-April poll, fielded in the days just after the speech, put his job approval at 83% — up five full percentage points from the 78% he was polling at just before the invasion (and the highest reading of his presidency; Gallup, 2016). Kennedy himself, looking at the numbers, reportedly told an aide: "The worse I do, the more popular I get."
He wasn't really joking. And he also wasn't really wrong.
A few years later, Elliot Aronson — who would go on to become one of the most-cited social psychologists of the twentieth century (as well as the author of my favorite book, The Social Animal) — would put a name to the phenomenon we had just observed. Aronson, Willerman, & Floyd (1966) recruited 48 college students and played them an audio recording of someone they were told was auditioning for a "College Quiz Bowl" team (think a trivia competition like Jeopardy!). The recording came in four versions:
- A superior candidate, who answers 92% of difficult questions correctly and is shown to have a stellar résumé.
- The same superior candidate, except, in the final moments, he audibly spills a cup of coffee on himself and exclaims, "Oh my goodness, I've spilled coffee all over my new suit!"
- An average candidate, who answers 30% of questions correctly and possesses a modest, if undistinguished biography.
- The same average candidate, but who now also spills the coffee on himself and makes the same exclamation.
The participants then rated each candidate on overall likability.
The results, in retrospect, may look obvious, but at the time they overturned a decade of assumptions about the relationship between competence and likability. The superior candidate who spilled the coffee was rated as the most likable of the four, while the average candidate who spilled the coffee was rated as the least likable. While each made the same blunder, the mistake itself had the opposite effect on likability. This is because, as it turns out, the act itself does not seem to universally affect how positively or negatively we evaluate someone; instead, the direction that the act pushes our evaluations in has a lot to do with how we perceive that person before they commit it.
The researchers named the phenomenon after a piece of theatrical slang: pratfall, a clumsy or sudden fall on the buttocks (the kind of thing that punctuates a vaudeville routine). The pratfall effect, in their words, was the counterintuitive increase in likability that a competent person gains when they reveal, in a small and harmless way, that they are also human.
The mechanism Aronson proposed was pretty straightforward: a person of obvious competence can begin to seem, in a way, inhuman — and, by extension, unrelatable. But a small, visible blunder can effectively collapse that distance and erase any uncomfortable feelings we might have towards those that seem "too perfect." In essence, these mistakes can help someone be seen as a person again (albeit a still very impressive one), complete with all the small dignities and indignities of one, rather than an icon.
But the average candidate didn't have that same distance to collapse; they were already seen as "just like us." In this case, the blunder didn't reveal anything redeeming about a person we'd already believed to be extraordinary; it simply confirmed what we already suspected about someone we'd already written off as ordinary. The coffee spill turned an icon into a (more likable) person, but it turned a regular person into a (somewhat cringeworthy) klutz.
Now, this is the part that almost every popular retelling of this research gets wrong. The pratfall effect isn't a simple parable about "being vulnerable." (That's not to say vulnerability isn't important — it is! — it's just too simplistic of an explanation for the results we see here.) Instead, it's a parable about how credibility can often benefit from being "humanized," but only after a certain level of credibility is earned.
When someone has a requisite level of perceived competence — when they're beginning to border on being seen as infallible — that's when showing a bit of fallibility can be a multiplier. In this way, we can think of the pratfall effect as more of a calibrated stumble. This is a bit more precise than mere blanket vulnerability, and it warns against the classic misreading of the results (i.e., "I guess I should just make more mistakes publicly!"). The goal isn't just to fail in public, it's to determine whether such a failure would improve your image (precisely because it was so "out of character") or simply cement it (because they already believed you weren't all-that-impressive to begin with).
The research since 1966 has only sharpened how conditional this effect is. Helmreich, Aronson & LeFan (1970) showed that the lift depends on who is watching (people with average self-esteem respond most strongly, while those with very high or very low self-esteem respond less). Mettee & Wilkins (1972) showed that severity matters (trivial blunders barely move the needle, while moderate, work-relevant ones seem to create real lift).
But there are a few very important caveats.
In 2008, across five studies, Brescoll & Uhlmann demonstrated that the same workplace blunder — expressing visible anger about a colleague's mistake — earned men higher perceived status and higher compensation but earned women lower status and lower compensation. This is critical as it suggests that the same act might be perceived differently (and thus have different consequences) based on identity characteristics such as gender. As frustrating as it may be, the risk/reward profile is likely not the same for a man and a woman; a highly-competent woman making a blunder might run more of a risk for their reputation (and see less of a humanizing "boost") than a highly-competent man making the same blunder (and I sincerely doubt this will come as a surprise to the women reading this newsletter, many of whom have plenty of experience navigating these frustrating workplace double-standards).
Additionally, last year, Nguyen, Artis, & Anderson (2025) showed that, across 350 B2B sales managers, managerial vulnerability — the deliberate willingness to admit mistakes and express uncertainty — strengthens coachability and lifts sales performance. However, it also temporarily lowers short-term learning effort (i.e., people might not work as hard doing things like learning new skills). For this reason, the authors call it a "double-edged leadership signal"; the stumble buys you enhanced trust and relational depth, but it might also come at costs elsewhere. In that way, you need to decide what is most important to you at the moment and what costs you're willing to endure to obtain it. Sadly, there are no free lunches here.
The lesson here isn't that leaders should pretend to be more flawed than they are. Instead, it's that competence and humanity (and all the fallibility it comes with) are not necessarily opposites in the eyes of the people watching us; they can be multipliers of one another (but usually only in that order).
When done in the right sequence, in the right room, by a leader the audience has already decided is firmly in command, the small human moment becomes one of the most powerful tools of influence we have. Done out of sequence, by someone who hasn't yet earned the floor, the same moment is just Dennis (predictably) spilling the coffee.
There are four places to put this research to work this month.
1. Earn the Floor Before You Break Any Windows. The pratfall premium is conditional on perceived competence. If you open a high-stakes presentation, a first board meeting, or a new client engagement with a blunder, you risk losing credibility before you begin. In other words, you have not yet earned the "right" (or necessity) to be humanized (because the room has not yet decided you were ever in danger of being super-human in the first place). The blunder, in that context, doesn't read as charming; it reads as unprepared.
2. Pick a (Small) Stumble That's About the Work. Some research has shown that the stumble that lifts a leader most is one that's about the work: a forecast you got wrong, a hire that didn't pan out, a question from the room you couldn't answer. But critically, this stumble has to come paired with the recovery: how you altered your forecasting models, changed your hiring criteria, prepared differently for important meetings. The blunder plus what you learned from it (and what you did about it) becomes a story, and people follow stories.
3. Calibrate to the Perceiver's Prior, Not Your Own Self-Image. You probably believe everyone in the room already knows you're competent, but the truth is, they might not. A new board member, a first-time audience, a skeptical journalist, a client you've never worked with before, these people are not in possession of the competence baseline that your team is. The calibrated stumble must be calibrated to them (and their perception of you). If the room has not yet seen the version of you that delivers, then it has not yet seen the version of you that gets to stumble. Adjust accordingly, either by earning the floor first or by saving the human moment for a room where the baseline already exists.
4. Beware the Asymmetry. The Brescoll & Uhlmann work — and a quarter-century of replications in its tradition — show that the calibrated stumble is not evenly distributed across the leaders in any given room. Women, junior executives, members of historically underrepresented groups, and anyone whose competence the audience may be (as unjustified as it might be) quietly questioning all face a smaller margin for the visible blunder to be seen positively. This isn't a reason to suppress humanity, but it is a reason to be especially discerning of when (and with whom) you use the technique. The same stumble that lifts one leader can land another in the file marked "Unprepared." The work of leadership — particularly for those whose authority, rightfully or not, isn't yet fully assumed — is to know which file the room is keeping on you, and to act accordingly.
Aronson, E., Willerman, B., & Floyd, J. (1966). The effect of a pratfall on increasing interpersonal attractiveness. Psychonomic Science, 4(6), 227–228.
Brescoll, V. L., & Uhlmann, E. L. (2008). Can an angry woman get ahead? Status conferral, gender, and expression of emotion in the workplace. Psychological Science, 19(3), 268–275.
Gallup. (2016, April 14). Gallup Vault: Cuba embargo popular after Bay of Pigs fiasco (news.gallup.com).
Helmreich, R., Aronson, E., & LeFan, J. (1970). To err is humanizing sometimes: Effects of self-esteem, competence, and a pratfall on interpersonal attraction. Journal of Personality and Social Psychology, 16(2), 259–264.
Mettee, D. R., & Wilkins, P. C. (1972). When similarity "hurts": Effects of perceived ability and a humorous blunder on interpersonal attractiveness. Journal of Personality and Social Psychology, 22(2), 246–258.
Nguyen, C. A., Artis, A. B., & Anderson, J. E. (2025). When sales managers stop pretending to be perfect: Vulnerability as a catalyst for sales coachability. Journal of Personal Selling & Sales Management (tandfonline.com).