How to help someone with impostor syndrome

Hannah comes home with a promotion and attaches an explanation to it before her coat is off: they filled the role in a hurry, she happened to be around, and nobody more suitable put their hand up. Ben says what almost anyone would say. "Nonsense, you're the best person they have." Hannah smiles, thanks him, and the subject is gone.

Three weeks later the same sentence about luck comes back, this time about a different project. Ben has heard it roughly ten times now, and it is dawning on him that his reassurance is not being filed anywhere. It does not sound hollow, he means every word of it, and it still does not survive the evening.

The sense of not belonging where you are was described by the psychologists Pauline Clance and Suzanne Imes in 1978, in people with documented results behind them. It is not an illness or a disorder and does not appear in the classifications of mental disorders. It is a way of interpreting your own successes, which is exactly why somebody close by has more leverage than it looks. Just not through praise.

Why reassurance bounces off

Doubt about your own ability does not rest on missing information. Someone who keeps thinking they have not earned their place usually knows what they have behind them, they just weigh it differently. Your compliment brings no new fact into that system, so it goes through the same filter as every compliment before it.

With people who are close, the filter runs faster still. Approval from a partner is explained away in a second: he says it because he loves me, because he can see I feel awful, because he does not know the details. An uncomfortable ratio is at work that most people do not expect. The closer you are to someone, the less your word weighs as evidence.

Mark Leary and his colleagues described in 2000 how large a part self-presentation plays in feelings like these. People talk about themselves more modestly than they rate themselves inside, partly in order to lower what the room expects. Your reassurance therefore often lands on a sentence that was half aimed at an audience, rather than on what the person actually thinks.

The second problem is quieter. When you tell somebody "that isn't true", you put them in a position where they have to defend their view, and a defence spoken out loud firms a position up rather than weakening it. The conversation meant to shrink the doubt gives it a voice and a set of arguments instead.

One more thing shows up only when you replay the exchange afterwards. The line "but you're brilliant" closes the topic. Nothing can follow it except "yeah, maybe", so a conversation that could have gone somewhere is over in two turns.

Three sources of doubt, three different conversations

Self-doubt does not have the same root in everybody. It is one pattern, but usually with one louder source, and what the other person needs to hear from you differs accordingly. The piece on what impostor syndrome is and where the doubts come from takes them apart in detail. For a conversation, this much is enough:

The difference between the columns is not kindness. The middle one holds judgments that can be argued with, the right-hand one facts and questions, which are far harder to argue with. Which of the three you hear most often at home or at work is easy to spot from the first column. The other side of the coin, the work done by whoever is receiving the praise, is covered in the piece on how to accept a compliment without the disclaimer.

What works instead of praise

The strongest tool is a dull one: description instead of a grade. "You're so capable" is an assessment, and assessments get explained away. "On Thursday you turned the whole meeting around by bringing in the numbers from three months back" is a description of an event the person was present for. Facts shrink less easily than compliments, and when they do get shrunk, at least there is something left to talk about.

Better still is to let the other person say the evidence out loud themselves. The question "what exactly did you do to make it land like that?" looks like polite interest, but it forces a reconstruction of the steps, and your own sentence cannot be dismissed as good manners. Silence helps here too: leave the pause longer than is comfortable, because the first answer tends to be a brush-off and the second one is worth having.

The third thing is normalisation, and paradoxically it carries the most weight. Clance and Imes worked with people who considered themselves the exception partly in groups, where it became audible that the person next to them knew the same thing. The large review by Bravata and her colleagues from 2020 pulled together 62 studies and found prevalence anywhere between 9 and 82 percent, depending on who was asked and how. The frequently quoted "70 percent" is an estimate by Sakulku and Alexander from 2011 rather than a measured figure, but either way we are talking about something very widespread.

The practical consequence is that your own doubts are more usable than your praise. When a manager tells the team he was not sure he was up to the role either, several people in the room breathe out at once. There is one risk: it must not be a run-up to the punchline about how brilliantly he handled it in the end.

A few sentences that work in all three cases:

• "What exactly did you do to make it turn out like that?"
• "I remember you saying the same thing three years ago about a job you then did for five years."
• Description instead of evaluation: "On Monday you handled a client who was unpleasant to everybody else."
• "Before every presentation I'm convinced I'm about to be found out. Every single time."
• "You don't have to agree with me. I just want you to know what I saw."
• The question that stings and helps at once: "What would somebody who deserves that position have to be doing? And which part of it are you not doing?"

Partner, friend, manager: three kinds of leverage

A partner is closest and, as a witness, unfortunately carries the least weight. What a partner does see is what nobody else does: how many extra evenings go into preparation, and how many opportunities get declined before they are ever mentioned out loud. That is the useful thing to name, because it is behaviour rather than a verdict. "I noticed you never actually signed up for that conference" opens a conversation that "you're good at this, you know" closes.

At home, one question at the start helps: do you want to solve this, or do you want to complain about it? Both are legitimate, and most arguments here start because one person offers solutions while the other wanted two minutes of being heard. You also do not have to be your partner's therapist, and attempts at it tend to cost the relationship before they cost the doubts.

A friend has the advantage of memory, because a friend knows the versions the person told about themselves five years ago and can quote them without sounding like a lecture. "You wrote me exactly this before you started your last job, and then you were the best person on that team" is not reassurance. It is data the friend was present for.

A manager or mentor has the most leverage, except that their praise goes through the filter too and comes out as "she was being nice to me". Assigning beats praising. A project handed over or a role given publicly speaks about trust in a way that cannot be explained away as politeness. Alongside that goes specific, regular feedback rather than one conversation a year at the review; how to give it so that it can be received is worked through in the piece on giving and receiving feedback.

It is worth saying that plenty of managers deal with the same thing themselves and simply do not mention it in front of the team. A promotion tends to amplify doubt, because the yardstick of expertise stops applying and a new one has yet to be found; there is a separate piece on impostor syndrome in a leadership role.

What to avoid

Comparison is the first trap. Leon Festinger described back in 1954 that we measure ourselves against others automatically, so there is no need to bring another yardstick into the conversation. "Look at Chris, he's in the same spot and doesn't make a drama of it" adds a second doubt to the first: now the way the person is reacting is wrong too.

The second trap looks friendly. "Everybody feels like that" is very nearly true as a fact, but as an answer it means "this isn't worth going into". The difference is audible at once: "everybody feels that, don't worry" closes the subject, while "I know that one, mine looks like this" opens it all the way. Same information, opposite effect.

The worst outcome comes from rescuing, usually from people with the best intentions. When you take a task over or leave someone off a more demanding project because "they're not in a good place for it right now", you confirm their doubt by an act. Words cannot refute it, and one protective move props it up reliably.

So, honestly: when did you last assure somebody that they are good, and when did you last describe to them what they specifically did? For most people the ratio between those two numbers is lopsided.

When to leave it and when to suggest a professional

Not every doubt is a problem to be solved. Mild doubt is familiar to almost everybody, it turns up before something new and then leaves again, and the best response to it is no particular response. If the other person does not want to talk, pushing turns the subject into something they have to defend, which is the opposite of what you intended.

It is different when the doubts last for months and start costing something. Declined opportunities, endless overtime, exhaustion or a mood that has been poor for a long time are signals your kind sentences will not cover. That is the point to say once that this is no longer about the two of you and your conversations.

An offer works better than a diagnosis. "I was wondering whether it might help to talk it through with somebody who works with this" is a sentence the other person can decide about on their own. Listing symptoms and naming what you think is wrong with them usually produces a defence. The decision is theirs anyway, and your job is to leave the door open rather than push somebody through it.

A conversation with nobody to convince

If you want a shared vocabulary for this, a shared measurement can serve. Take the impostor syndrome test both of you and compare where your strongest source of doubt comes out. The result is an approximate picture rather than a verdict, but it makes a decent opening for the sentence "yours runs mostly through circumstances, mine through playing things down", which would sound like an accusation any other way.

Try one small thing before the next time. When the line about luck or coincidence or it being nothing arrives, do not answer with disagreement. Ask one question about the specific steps, then stay quiet for ten seconds. Do nothing else that evening.

Whether it worked will not be visible that day. It will be visible in a month, from one thing only: whether the other person brings it up again on their own, or whether they keep it to themselves after all.