I Assumed You’d Say No. I Never Checked.

The Load  ·  The Reading Room, Vol. 18  ·  9 min read  ·  What twenty years of compliance research found about who says yes, why the refusal you are hearing was never theirs, and the one small ask worth testing this week

Somewhere along the way you became the reliable one, and asking for help when you’re the capable one started to feel like breaking a rule nobody ever actually wrote down.

Two Saturdays ago I wedged a queen mattress sideways on my own staircase landing, alone, on purpose, because asking my nephew to come down one flight of stairs felt like a bigger imposition than dislocating a shoulder.

He was in the house. Two rooms away, watching something with a laugh track, fully capable of holding one end of a mattress for ninety seconds.

I did not ask him. I fought a mattress like it had said something about my mother.

At work I’m the person people quietly copy on an email so there’s a written record, and the person they call when they don’t want it written down yet. Both are a form of trust. Neither has made me better at asking for something myself. What follows is a reading of published social psychology on that failure; I am not a clinician, the findings belong to the researchers named below, and only the plain-English version is mine.

The short answer: the usual explanations do not hold up. Pride would mean struggling visibly rather than admitting you need something. Martyrdom would mean enjoying the carrying. Neither is what is happening. What is running underneath is quieter and more mechanical, and it has a name worth having: the borrowed no. You issue a refusal on the other person’s behalf, in their voice, without their knowledge, and then you obey it. Twenty years of compliance research says that borrowed no is wrong far more often than it is right.¹²

I know exactly why you won’t ask

I know exactly why you won’t ask, because I have watched myself do it for years and given it a dignified name. Independence, mostly, or not wanting to make things harder for someone else.

Neither is quite a lie. Both are the story the mechanism tells about itself afterwards, which is how an error ends up sounding like a virtue.

Pride would predict struggling visibly rather than admitting you need something. You do the opposite — struggling invisibly, at eleven at night, over a flat-pack wardrobe with the instructions in the wrong language. Martyrdom would predict some private satisfaction in the suffering. There is not any. Just arithmetic you were never taught to question, and the same wrong answer every time.

I have already written about the two words behind half of this, the reflexive “I’ve got it” that leaves your mouth before you have decided anything. Asking for help lives on the other side of that same reflex, unused.

Here is the actual math behind asking for help when you’re the capable one

None of this happens in a vacuum. You are likely already holding the four-part job researchers can now measure — anticipating, identifying options, deciding, monitoring — for people who have no idea it is four jobs rather than one. A fifth job, asking, is where the arithmetic breaks down.

Definition: the underestimation-of-compliance effect is the reliable gap between how many people an asker predicts will say yes and how many actually do. It is measured by having people forecast a number before they ask, then counting.²

Francis Flynn (Stanford GSB · site) and Vanessa Lake — who publishes today as Bohns — ran six studies to measure it. In the simplest, participants had to get five strangers to fill in a short questionnaire, and guessed beforehand how many people they would need to approach.

They guessed wrong, and wrong in the same direction, every time. Across the six studies, including real charity fundraisers going door to door, 79 percent of the people doing the asking underestimated how many people would say yes.¹

The reason is not mysterious. Askers carefully price what a favour costs the other person — the time, the errand, the logistics. What they never price is the discomfort of refusing a direct request, to someone’s face. That discomfort does most of the work of getting you a yes, and it has never once made your ledger.

The problem was never your pride. It was your sum, and nobody ever taught you to check it.

If you want to see the whole list of asks you have quietly run this same bad sum on — every job you are carrying that nobody has actually agreed to — the Quiet Audit is a private ten-minute pass through it. No programme, no advice, nobody sees your answers. Start here.

The refusal you are hearing is one you wrote yourself

There is a version of this that sounds like generosity: you do not want to burden anyone. Fair enough. But look at what it actually does.

You decide, alone, on their behalf, that the ask is too big, too awkward, too much. They never got a vote. You ran the whole negotiation in your head and lost it to an opponent who was never in the room. You cross-examined a version of your sister that exists only in your head, and she never once got to disagree with the verdict.

The two refusals are not the same object, and it is worth seeing them side by side.

Somewhere in your life is a person who would say yes in the time it takes to read this sentence, and might be glad you asked. You have never given them the chance. That is not humility so much as deciding, alone, that you already know the ending to a conversation you never had.

Five Things That Look Like This and Are Not This

Everything above assumes the people around you are ordinarily willing and the problem is your arithmetic. Sometimes that is wrong, and “just ask” is the wrong instruction to give a woman in any of these five situations.

• Asking has actually been punished. If requests in your house have been met with retaliation, a sulk that lasts three days, or an invoice presented weeks later, you are not miscalculating. You have learned an accurate rule about a specific person, and the question is about that relationship rather than your confidence.
• Social anxiety as a condition rather than a habit. If the dread of being judged shows up across most situations, not only when you need something, and has narrowed what you are willing to do, that is a recognised and treatable pattern — and worth naming to a doctor rather than solving with a better script.
• Depression’s withdrawal. “I will manage” and “nothing anyone did would help” are different sentences with different mechanisms. If the flatness is there on a genuinely clear weekend too, the responsiveness test is the more useful place to start.
• A roster that genuinely is not there. After a bereavement, a move, or a stretch where everyone you know is carrying something, the problem is not your estimate. It is the list. Rebuilding a roster is slower work than one ask and does not respond to this article.
• A workplace where asking is genuinely scored against you. Some are. If you have watched a request cost you standing twice, that is evidence rather than distortion, and the answer is a record and an ally rather than more confidence.

None of these means the arithmetic problem is not also real. They mean the test below is the cheap first move, not the whole answer, and if it changes nothing in a fortnight, the list above is where to look next.

Who Else Has Measured This

One study names an effect; it takes other work to show it holds. Three lines converge here.

The channel research. M. Mahdi Roghanizad (Toronto Metropolitan Univ. · Scholar · LinkedIn) and Bohns had people ask friends to proofread half a page over different channels, and compared what askers predicted with what actually happened across roughly 1,490 people asked. Face-to-face beat email and text; video and phone sat in between.³ Askers thought what mattered was what they were asking for. What mattered was how they asked.

The effort research. Daniel A. Newark (HEC Paris · Scholar), Bohns and Flynn, writing in Organizational Behavior and Human Decision Processes in 2017, found help-seekers also underestimate how much effort helpers actually put in once they say yes — which matters here, because the fear is usually not just of refusal but of a grudging half-yes.

The discomfort research. Bohns and Flynn’s earlier work in the Journal of Experimental Social Psychology found people overestimate how awkward help-seeking will feel, in the same direction and for the same reason: the imagined scene is harsher than the real one.

A set of field experiments that found the effect, a review that priced it across fourteen thousand requests, and follow-up work showing the same error turns up in the channel, the effort and the awkwardness. The convergence is on the direction of the error, not the size of any single number.

The rehearsal is almost always longer than the conversation.

What This Actually Changes

Not your character. Nobody is coming to take away the fact that you are the capable one, and you would not hand it over if they did.

What it changes is the standing of the verdict. You have been treating “they would rather I did not ask” as something you know. It is a forecast, made by an instrument with a documented direction of error, and it has never once been checked against the thing it claims to predict.

It also changes the order of operations. The instinct is to wait for an ask big enough to justify the awkwardness — the appointment, the deadline, the week you cannot cover. That is the worst possible test case, because the stakes make a refusal expensive and you will not run it. Test the cheap thing first, on something you could survive being wrong about.

What to Say, and What You Will Probably Hear Back

Not the mother’s appointment. Not the deadline. Start smaller than feels reasonable, because the smallness is the point — you are testing the arithmetic, not solving your life in one afternoon. And say it in person or on the phone if you can, because that is the condition under which the research says the yes actually arrives.³

Notice what those asks have in common. Each one is specific, small, and answerable in a word. Vagueness gives the other person a project to scope rather than a question to answer, and a project is much easier to decline.

Questions people ask me about this

Why is asking for help when you’re the capable one so hard?
Because the capable one gets no practice at it, and because the sum is wrong in a specific way. Askers price what the favour costs the other person and never price how uncomfortable it is to refuse a direct request. Across six studies, 79 percent of people asking underestimated how many would say yes.¹

Isn’t this just lowering my standards for who I rely on?
No. Lowering your standards means asking someone unreliable and hoping. This is the opposite: asking someone capable who was never given the chance, and letting them answer on their own evidence instead of your forecast.

What if I ask and the answer actually is no?
Then you have learned something real rather than something invented. The studies measured refusals too — just less often than people expected. One real no costs far less than the thousand imagined ones you have been issuing on their behalf for years, none of which they ever knew about.

Does the way I ask make a difference, or only what I ask for?
Both, and the channel matters more than most people assume. In a study of about 1,490 requests, face-to-face asks were granted at a markedly higher rate than the identical request by email or text, and askers did not anticipate the difference.³

Does this apply at work the same way it applies at home?
Closely enough to be worth testing in both. Much of the original research is workplace-adjacent — colleagues, acquaintances and near-strangers doing each other favours. The exception is a workplace where asking has genuinely been scored against you, which is a real situation and is in the differential above.

Where to Go Next

In order, and each for a reason.

1. The reflex that says “I’ve got it” before you have decided anything — the front half of this, and the cheaper place to intervene.
2. Then why you resent the people you chose to help, which is what an unasked question turns into after a few years.
3. Then why you read a two-word reply four times, if the pre-refusing is really about what you think people already think of you.

It was true for me and the mattress in the end. The asking would have taken four words. The not-asking took twenty minutes and a mild strain in my shoulder.

References
1. Flynn, F. J., & Lake, V. K. B. (2008). If You Need Help, Just Ask: Underestimating Compliance With Direct Requests for Help. Journal of Personality and Social Psychology, 95(1), 128–143. Read the study
2. Bohns, V. K. (2016). (Mis)Understanding Our Influence Over Others: A Review of the Underestimation-of-Compliance Effect. Current Directions in Psychological Science, 25(2), 119–123. Read the study
3. Roghanizad, M. M., & Bohns, V. K. (2021). Communication channel and predicted versus actual compliance. Social Psychological and Personality Science. Cornell summary of the findings

On the researcher. Vanessa Bohns is a social psychologist and organisational behaviour scholar; she is not a clinician, she is not your clinician, and nothing in her published work is advice about your family. She has written two trade books on this subject — You Have More Influence Than You Think (W. W. Norton, 2021) and Should I Say Something? (Hachette, 2026) — which is a commercial interest in the topic and is disclosed here rather than left for you to find. The studies are linked in full above, and the sample sizes and dates are stated in the body rather than buried.

Written by Jo Brennan for Blue Leaf Journal. Updated: 28 August 2026.