People tend to brace for a harsher reaction to their no than they actually get. In a 2024 series of experiments, people who declined an invitation expected the person who asked to be angrier, more disappointed and less likely to invite them again than people on the inviting side reported.¹ The worry is real, but it runs ahead of the other person’s reaction.

So how to say no at work comes down less to a perfect phrase than to two things: deciding before you answer, then saying it plainly and warmly. Saying no is one habit within managing your time well; this guide stays with the refusal itself. Six steps, from deciding to delivering:

1. Ask for time before you answer
2. Check the request against what you have already promised
3. Say no early, plainly and kindly
4. Give a short reason that is true
5. Offer a smaller yes only if you mean it
6. Write your no before you need it

Saying no does less damage than you predict

Declining usually does less harm than the person declining expects, according to experiments by Julian Givi of West Virginia University and Colleen Kirk of New York Institute of Technology, published in the Journal of Personality and Social Psychology in 2024. People on the inviting side reported being less upset than the decliners imagined.

The study

Moderate evidence

Does a no sting as much as the decliner thinks? Five experiments, 2024

Participants imagined either turning down an invitation, such as a friend’s offer to visit a museum exhibit, or being turned down. Decliners rated the fallout as worse than inviters did: more anger, more disappointment, more harm to the relationship and a lower chance of future invitations. The gap held in a study where real couples wrote and declined real invitations, with the decliner told to say no. In a pilot survey for the paper, 77% of respondents said they had accepted an invitation they did not want because they feared what saying no would bring.¹

The gap was small: decliners expected somewhat more hurt, not a different relationship altogether. The main caveat is scope. The invitations were to everyday fun activities, most scenarios were imagined, and the authors leave invitations to useful rather than fun activities for later studies.¹

A separate team found the same pattern for requests. In preregisteredpreregistration: Filing a study's hypotheses and analysis plan publicly before the data are collected or examined. It stops a planned test from being rewritten once the results are in, so a reader can tell a real confirmation from a hunt through the data.Full entry in the glossary studies published in 2023, Jingyi Lu and colleagues at East China Normal University found that people who refused a request overestimated the negative consequences they would face, whether the person asking was a friend, a colleague or a stranger. Most requests were small, and the participants were recruited in China.²

Why the mismatch? Givi and Kirk’s evidence points to attention. When you say no, you know how much you weighed it: the pull to agree, the reasons you could not. You picture the other person seeing only the bare refusal. In their studies, people in the inviter’s role gave more weight to the thinking behind the no than decliners expected.¹ Lu’s team points to a self-protective motive instead: people overrate the fallout partly because they badly want to avoid it.² Givi and Kirk’s data did not support that motive for invitations, so the two teams explain the gap differently.¹

• The refusal itself
• The thinking behind it

1. What the person saying no pictures: the bare refusal fills the frame; the thought behind it barely shows
2. What the person turned down tends to weigh: the refusal still registers, but so does the thought that went into it

Picture a colleague inviting you to a long team lunch on the day a report is due. You imagine her remembering the refusal for weeks. If the finding holds in offices, she is more likely to register that you considered it, had a reason and said so kindly. Givi and Kirk found one simple correction: when decliners first imagined being the one who asked, their predictions came into line with the inviters’ views.¹

So before you accept something just to avoid a reaction, picture yourself making the request and hearing a polite no, then decide.

The awkwardness of refusing pushes people to say yes

Refusing feels so uncomfortable that people asking for help underestimate how often others will agree, according to 2008 studies by Francis Flynn and Vanessa Bohns (published as Vanessa Lake). Askers weighed what saying yes would cost the other person and overlooked how awkward it is to say no.³

In their first field test, students asking strangers on a US campus to fill in a questionnaire expected to need about twice as many approaches as they actually did. Later studies in the same paper pointed to a reason: people imagining themselves as the asker rated a refusal as less awkward and embarrassing than people imagining being asked.³

Read from the other side, this explains a familiar moment. A teammate stops by late in the day and asks you to review a slide deck before tomorrow’s pitch. You feel a strong pull to agree, and it can seem like a sign that the request matters. Often it is simply the social cost of refusing, which the teammate probably cannot see.

A yes given under that pressure has costs of its own. Summarizing earlier research, Rachel Schlund, Roseanna Sommers and Vanessa Bohns note in a 2024 paper that people who feel guilted into agreeing can feel regret and resentment, and may give lower-quality help or back out later.⁴

So separate two questions that the moment blurs: is this awkward to refuse, and is it worth my time? Only the second should decide your answer.

Managers can use the same finding from the other side of the desk. In two preregistered lab experiments with US university participants, the 2024 study found that giving people the actual words to decline, rather than only telling them they were free to refuse, made them feel freer to say no or yes. The request was unusual (handing over an unlocked phone), the effect showed up in feelings rather than clearly in behavior, and differences in power were not tested.⁴ Even so, when you hand someone a task, adding a ready-made way out, such as “If this week is full, just tell me ‘not this time’,” takes one sentence and may make an honest answer easier to give.

How to say no at work, from first ask to final answer

These six steps apply the research above. The evidence for each is indirect, so treat them as reasoned habits rather than tested techniques.

1. Ask for time before you answer

A request arrives now, but its cost lands later, and later always looks roomier. In seven experiments published in 2005, Gal Zauberman and John Lynch found that people expect to have more spare time in the future than they have today, an expectation more pronounced for time than for money.⁵ That helps explain why a yes for next month is easy to give and hard to keep.

“Let me check my week and get back to you this afternoon” is a complete answer. It also gives you room for a quick test.

2. Check the request against what you have already promised

Put the request next to this week’s commitments and name what would move if you agreed. Ranking tasks by what they are worth, one of the starter habits of good time management, makes that comparison quick.

Notice, too, which kinds of requests keep reaching you. In experiments reported in the American Economic Review in 2017, women were asked more often than men to take on a task everyone preferred someone else did, and agreed more often, partly because people expected them to.⁶ When the task matters but someone else is better placed to do it, the better answer may be to delegate rather than decline.

3. Say no early, plainly and kindly

A late no leaves the other person less time to find someone else, so answer as soon as you know. Lead with the answer, then the warmth: “I’m going to pass on this one. Thanks for thinking of me.” If the thought you put into the decision is what people tend to notice, let it show in one sentence rather than an apology spread over a paragraph.¹

4. Give a short reason that is true

In Givi and Kirk’s experiments, decliners gave a plain reason, wanting to stay home and relax, and still overrated the fallout, so the reason does not need to be impressive.¹ Keep it short and tie it to something the other person can recognize: “I’ve promised the client draft for Thursday, and this would put it at risk.” If the real problem is the deadline rather than the task, negotiating the date is a different conversation from refusing.

5. Offer a smaller yes only if you mean it

A counteroffer, such as half an hour instead of the whole review, next week instead of Friday, or the name of someone who knows the system, keeps things warm only if you deliver it. Offered just to escape the awkwardness, it is the reluctant yes in smaller form, and it can breed the same resentment.

6. Write your no before you need it

Part of what makes refusing hard is finding the words when you are put on the spot, the 2024 Schlund, Sommers and Bohns paper argues.⁴ Draft two or three sentences you are comfortable saying for the requests you get most often, such as extra meetings, reviews and late messages, and keep them somewhere you can find them.

Change the wording until it sounds like you, because a script you would never say out loud will not help in the moment. The same thinking applies to your calendar: protected focus blocks are easier to defend when you have already decided what they are for.

WiserHours

When the next request lands

• Ask for time instead of answering on the spot
• Ask whether you would still say yes if it were due next week
• Name what the yes would push out
• Answer early, lead with the no, then thank them
• Give one short, true reason
• Offer a smaller yes only if you will deliver it
• Keep two or three ready-made refusals in your notes

Find this checklist online:
https://wiserhours.com/time-management/say-no-at-work/#checklist-say-no-at-work-steps

“I don’t” or “I can’t”: what the wording research found

The popular advice to say “I don’t” rather than “I can’t” traces to two small sets of experiments by Vanessa Patrick and Henrik Hagtvedt, both published in 2012, and our search turned up no independent replicationreplication: Running a study again with fresh data and the same method, to see whether the original result comes back. A finding that fails to replicate is not automatically wrong, but it stops being something you can lean on.Full entry in the glossary for either paper. The wording may help, but it should not carry the refusal on its own.

The better-known paper, a Journal of Consumer Research study, was about self-talk, not colleagues: people resisting a tempting snack or sticking to a personal goal did better when they told themselves “I don’t” than “I can’t.” Its studies were small, the largest with 120 students.⁷ A companion paper tested refusals aimed at other people. In a role-played sales pitch and an imagined party where a guest turned down cake, a refusal framed as “I don’t” came across as more convinced and more persuasive, across fewer than 100 people in its two main studies.⁸

The authors’ explanation is worth knowing even without a replication. “I can’t” names an obstacle, and obstacles invite discussion and negotiation (“What if I take your Monday meeting?”), while “I don’t” names a settled choice. They add their own warning that there is a fine line between firm and rude, and suggest that firm but polite refusals work best.⁸

Research on assertiveness points the same way. In 2007 studies built on coworkers’ comments and ratings of colleagues, Daniel Ames and Francis Flynn of Columbia University found that people seen as highly assertive were linked to strained relationships, and people seen as unassertive to getting less done.⁹ Both extremes cost something, which is why assertive communication at work sits in the middle.

So pick the words that are true for you. “I don’t take on new reviews in the last week of the quarter” works if it is a real rule; “I can’t” is honest when the obstacle is real. Either way, say it once, calmly, and let it stand.

Which refusal findings hold up best?

The best-supported finding here is that people overestimate the social cost of saying no, seen in a 2024 set of experiments on invitations and a 2023 set on requests. The advice on wording rests on small studies without independent replication, and no study followed real working relationships over months.

In practice, lean on the best-supported point, that the reaction you fear is usually milder than you expect, and treat any rule about exact wording as a preference rather than a finding.

The bottom line

Most of the fear around saying no is a forecast, and the research suggests the forecast runs pessimistic. Take a moment before you answer, say no early and kindly with one true reason, and keep the wording simple. The person you turn down is likely to notice that you thought about it, not only that you declined.