.png)
Boundaries That Stick: What Makes a Line Hold in a Close Working Partnership.
Updated: Sep 28
Three months ago your executive asked whether you could take the minutes at the leadership team meeting, only this once, because the person who normally does it was on leave. You said of course, because it was one meeting and you were going to be in the room anyway. She came back in the spring. You are still taking the minutes, and last week somebody asked whether you could also chase the actions afterwards, since you have the list.
Nobody decided that. Nobody sat down and agreed that the leadership minutes had become part of your job. It happened the way most of these things happen, one reasonable yes at a time, until the favour became the arrangement and the arrangement became the job. I wrote about boundaries that stick in 2025 and opened with a confession I will repeat here, because it is still true: I did not start my career with brilliant boundaries. I started it with stamina, colour-coded systems and a quiet belief that if I absorbed enough of the chaos, everyone around me would finally feel lighter. They did. I did not. This piece is about the part that comes after the line is drawn, which is what makes a boundary hold in a close working partnership, and what quietly moves it.
What you will get out of this
The study of 431 managers and their direct reports in four countries that found the two sides of the desk draw the edges of the same job in different places, and which side draws them wider
Why the arrangement between you and your executive is a contract whether or not anyone wrote it down, and what happens to the half of those contracts that get broken
The one kind of negotiated arrangement that came with more unpaid overtime rather than less, and what to ask for alongside it
How a favour turns into a job, from the foot-in-the-door experiments to a study of 3,680 employees whose extra effort cost them pay rises
The finding that dependable people are handed more work and credited with less effort, by the very people relying on them
The line-crossing illusion: why the moment you feel you have pushed too hard is usually the moment you got it right
Why being predictable is kinder than being lovely some of the time, and what to do in the week a boundary slips
The executive's side of the line, and what power does to how clearly a leader sees the load beside them
The boundary check: a tool that reads one thing that has crept into your role and tells you how to put it back on the table
The Boundary Pages: three pages for drawing the edges of your role, agreeing them out loud and keeping them
Where your job ends, and who decides
Every close working partnership has an edge, the place where your job stops and something else begins. The trouble is that the edge exists in two heads, and the research is very clear that it is rarely in the same place in both.
In 1994 a study of 317 clerical workers asked people which behaviours they considered part of their job and which they considered going beyond it. The answers varied a great deal from person to person, even among people doing the same work, and it tracked how committed people felt and the cues they picked up around them. The same helpful act was a duty to one person and a favour to the next.
Five years later a larger study paired 431 supervisors with one direct report each, across the United States, Australia, Japan and Hong Kong, all in the same jobs in the same company. Both sides rated whether a list of helpful, conscientious, going-the-extra-mile behaviours was an expected part of the direct report's job. The supervisors consistently drew the job wider than the people doing it. Culture mattered too: participants in Hong Kong and Japan were more likely to see some of those behaviours as simply part of the job.
That finding explains a great deal of quiet resentment in this profession. If you think of rearranging the leadership offsite around a board member's flight as going above and beyond, and your executive thinks of it as the job, then every time you do it you are giving a gift that is being received as a delivery. Nobody is being unkind. You are each reading from a different map.
The maps also shift with seniority. A study of 469 workers found that helping and taking part in the wider life of the organisation were more likely to be seen as part of the role for more senior employees than for junior ones. The better and more established you get, the more of your generosity is quietly reclassified as your job. And the researchers who first defined going beyond the job eventually conceded the point themselves: by 1997 the leading figure in the field argued that it was no longer useful to treat those behaviours as extra, beyond the job or unrewarded, because in practice they rarely were.
Where the edge is blurry, people pay for it. Role ambiguity, meaning not knowing what is expected of you, was one of the first things organisational researchers learned to measure, and a meta-analysis of 169 samples and more than 35,000 employees found it among the stressors most strongly linked to poorer job performance. A separate meta-analysis of 42 studies found that role ambiguity and role conflict went with less helping, not more, while being overloaded went with a little more of it. That last detail is worth sitting with. People who are already stretched keep helping. They do not stop to renegotiate; they absorb.
The same favour can be a gift to the person giving it and a delivery to the person receiving it, and neither of them is wrong about what they meant.
There is a particular kind of work where the maps diverge most, and it is the emotional kind. A study inside a recruitment agency looked at managers who helped their staff through difficult feelings. The managers saw that help as over and above their duties and hoped for some commitment in return. The staff saw it as part of what a manager is for, which needed no reciprocation at all. Both sides accepted that the help mattered. The managers ended up disappointed, and the staff never knew there was anything to be disappointed about. If you have ever been the person who absorbs the frustration before a board meeting, reads the tone of an email so your executive does not have to, or calms the room that everyone else stirred up, you will recognise the shape of that.
Much of this work is also invisible by accident of design. A 2013 survey of 1,011 secretaries in the UK found a profession that was better qualified and generally better paid than twenty years earlier, with a minority doing complex managerial work, and with career prospects that remained bleak across the board. Sociologists describe how cultural, legal and spatial arrangements make some labour disappear from view, and a study of 185 interviews with social service workers found that informal extra work became invisible in layers: to the organisation, to society, and eventually to the workers themselves. Invisible to the self is the layer that matters here. You cannot defend an edge you have stopped seeing.
Try this. Take a sheet of paper and draw three columns: what I own, what I support, and what is not mine. Fill it in for your role as it is today, not as it was advertised. Then, if the relationship can hold it, ask your executive to fill in the same three columns separately, without seeing yours. The comparison will not be comfortable, and it will be the most useful twenty minutes you spend this quarter, because every difference between the two sheets is a place where a boundary is currently being held by one of you and not the other.
The research behind this section
Morrison (1994) surveyed 317 clerical workers and found that employees differed in what they defined as in-role and extra-role behaviour, and that these differences were related to commitment and social cues. Lam, Hui and Law (1999) studied 431 independent supervisor and subordinate pairs in the United States, Australia, Japan and Hong Kong, all in the same jobs in the same company, and found that supervisors had broader definitions of the subordinate's job role than subordinates did, and that participants from Hong Kong and Japan were more likely to regard some categories of citizenship behaviour as an expected part of the job. Turnipseed and Rassuli (2005) found that managers and employees perceived citizenship behaviour differently, with managers rating employees lower than employees rated their peers, and recommended that managers clarify which behaviours are extra and which are expected. Kang and Ryan (2016) surveyed 469 workers across countries and found that helping and civic virtue were more likely to be regarded as in-role behaviour for more senior employees than for junior ones. Organ (1997) argued that it was no longer fruitful to regard citizenship behaviour as extra-role, beyond the job or unrewarded by the formal system. Rizzo, House and Lirtzman (1970) developed measures of role conflict and role ambiguity and found them related to satisfaction, anxiety and propensity to leave. Gilboa, Shirom, Fried and Cooper (2008) meta-analysed 169 samples of 35,265 employees and found role ambiguity and situational constraints most strongly negatively related to job performance among seven work stressors. Eatough, Chang, Miloslavic and Johnson (2011) meta-analysed 42 studies and found role ambiguity and role conflict negatively related to citizenship behaviour, with a positive direct relationship between role overload and citizenship behaviour. Toegel, Kilduff and Anand (2013) found in a recruitment agency that managers regarded helping employees with negative emotions as over and above their duties and expected reciprocated commitment, while employees defined it as part of the managerial role requiring no reciprocation, leaving managers disappointed at the lack of reciprocity. Truss, Alfes, Shantz and Rosewarne (2013) surveyed 1,011 secretaries and found them better qualified and generally well paid, with a minority undertaking complex managerial tasks, most performing traditional tasks, and career prospects remaining bleak. Star and Strauss (1999) set out the ecology of visible and invisible work. Hatton (2017) identified cultural, legal and spatial mechanisms that make work invisible and contribute to its devaluation. Lavee and Kaplan (2022) analysed 185 in-depth interviews with social service workers and found that informal practices became invisible as work in several layers, to the self, the organisation and society.
Making the deal out loud
Every close working partnership runs on a deal. Most of them are never written down, which does not make them any less real.
Researchers call it the psychological contract: your beliefs about what you owe the other person and what they owe you in return. It forms in the first weeks, out of what was said at interview, what happened the first time something went wrong, and a thousand small precedents nobody noticed setting. One of the architects of the idea later described its building blocks as schemas, promises and mutuality, and the last of those is the one that matters most here. A deal only works when both people believe they have made the same one.
Here is how often they do not. In a study of 128 business school graduates surveyed at the start of their jobs and again two years later, 54.8% said their employer had broken the psychological contract. Those breaches went with lower trust, lower satisfaction and more intention to leave. A daily diary study found that broken promises, and exceeded ones, turned up regularly and in relation to almost any part of work, and that how much the promise mattered to the person decided how strongly they felt it. A meta-analysis confirmed that breach was related to almost every work outcome it examined, from commitment and satisfaction to performance and helping, and that it worked through feelings of violation and mistrust. The one exception was actual resignations, which did not follow as reliably as the intention to leave did. People stay, and they stay unhappily.
The encouraging evidence comes from pairs who do agree. A study of 80 pairs of research directors and the staff scientists who worked for them measured how closely the two sides' versions of the deal matched. Where they matched, and where each side felt the other was holding up their end, the scientists were more productive on archival measures, advanced further in their careers, felt their expectations had been met and wanted to keep working together. A meta-analysis of 64 samples and 10,884 leader and follower pairs found that the two sides' views of their relationship agreed only moderately overall, and agreed more the longer they had worked together and the more often they interacted.
That gives you the first rule of boundaries that hold: agreement is built, not assumed, and it is built by talking about the deal itself rather than about this week's tasks.
The research on idiosyncratic deals, the personal arrangements individuals negotiate that differ from the standard terms of their role, shows what that talking can achieve, and one trap inside it. A survey of 887 employees in a German public agency found that people who took initiative were more likely to negotiate personal arrangements, and it split those arrangements into two kinds. Deals about flexibility, such as when and where people worked, were linked to less conflict between work and home and less unpaid overtime. Deals about development, such as a new responsibility or a chance to grow, were linked to higher performance expectations and more commitment, and also to more work-family conflict and more unpaid overtime.
Read that twice if the next conversation you are planning is about a bigger remit. The growth deal is the one that carries the hidden cost. When you negotiate the new responsibility, negotiate what goes to make room for it in the same conversation, because the research suggests nothing will leave by itself.
A study of 263 supervisors added the view from the other side. They granted development deals mainly to people who showed initiative, granted flexibility according to the kind of work involved, and granted reductions in workload mostly where they felt the organisation had already fallen short of its obligations. In other words, asking to carry less was most likely to succeed when both people could see the deal had become lopsided. A three-wave study of 230 employees and 102 supervisors found that personal deals raised proactive behaviour and commitment, both because the relationship felt like a fair exchange and because the deal made people feel valued.
Two more findings make the case for doing this early and out loud. In a study of 32 teams, those that invested early in both a charter for how they would work together and a strategy for the work sustained the highest performance. And a meta-analysis of 52 studies and roughly 17,000 people found that realistic previews of a job reduced people leaving mainly by signalling that the organisation was being honest with them. Saying plainly what the role includes and excludes is not a cold act. It reads as honesty, and honesty is what people repay.
The first weeks of any new pairing set more than you would think. A study of how norms form in new groups found that people arrive with scripts from similar situations, act on them, and either tacitly go along with the direction the group is taking or challenge it openly. Silence was a vote. Every week you do the thing without comment, you are agreeing to it.
Try this. Write a one-page working agreement with your executive, and give it a review date. Four headings are enough: what I own, what I support, what we agreed is not mine, and how we define urgent. Add a line for the next development opportunity and what will move off your plate to make room for it. Then put the review in both diaries, three months out, and treat it as a normal part of the job rather than a crisis meeting.
The research behind this section
Rousseau (1989) set out the idea of psychological and implied contracts in organisations, and Rousseau (2001) described schemas, promises and mutuality as the building blocks of psychological contract formation. Robinson and Rousseau (1994) surveyed 128 management graduates at recruitment and two years later and found that 54.8% reported their psychological contract had been violated, with violations correlating positively with turnover and negatively with trust, satisfaction and intentions to remain. Conway and Briner (2002) used daily diaries and found that broken and exceeded promises occurred regularly in relation to virtually any aspect of work, and that the importance of the promise contributed significantly to emotional reactions. Zhao, Wayne, Glibkowski and Bravo (2007) meta-analysed psychological contract breach and found it related to all eight work outcomes examined except actual turnover, with the effects mediated by feelings of violation and mistrust. Dabos and Rousseau (2004) studied 80 employee and employer pairs in 16 university research centres and found that mutuality and reciprocity in their psychological contracts were positively related to research productivity, career advancement, met expectations and intention to continue working together. Sin, Nahrgang and Morgeson (2009) meta-analysed 64 samples of 10,884 leader and member pairs and found moderate agreement about the relationship (a corrected correlation of .37), which was higher with longer relationship tenure and, in a further 98 pairs, increased with the length and intensity of interaction. Hornung, Rousseau and Glaser (2008) surveyed 887 employees in a German government agency and found that personal initiative predicted negotiating idiosyncratic deals; flexibility deals were negatively related to work-family conflict and unpaid overtime, while developmental deals were positively related to both, as well as to increased performance expectations and affective commitment. Hornung, Rousseau and Glaser (2009) surveyed 263 supervisors and found that they authorised developmental deals in response to employee initiative and granted workload reduction deals in the context of unfulfilled organisational obligations. Liao, Wayne and Rousseau (2016) reviewed the research and meta-analysed 23 studies of 8,110 people. Liu, Lee, Hui, Kwan and Wu (2013) studied 230 employees and 102 supervisors over three waves and found that idiosyncratic deals raised proactive behaviour and affective commitment through both social exchange and self-enhancement. Lai, Rousseau and Chang (2009) studied 65 employees in 20 research and development groups and found coworkers more accepting of a colleague's idiosyncratic deal when the two were close friends and when they believed a comparable opportunity would come to them. Ashford and Black (1996) found that newcomers used job-change negotiating among other proactive tactics during organisational entry. Mathieu and Rapp (2009) found in 32 teams that those high on both charter quality and performance strategy quality sustained the highest performance over time. Earnest, Allen and Landis (2011) meta-analysed 52 studies with roughly 17,000 people and found that realistic job previews influenced voluntary turnover mainly through perceptions of organisational honesty. Bettenhausen and Murnighan (1985) studied 19 new decision-making groups and found that members drew on scripts from past experience and either tacitly accepted the direction the group was taking or openly challenged the implied norm.
How a favour becomes a job
Boundaries rarely break. They drift, and the mechanism behind the drift has been studied for sixty years.
In 1966 two psychologists showed that people who had agreed to a small request were much more likely to agree to a larger one afterwards, and the effect held even when a different person made the second request. They called it the foot in the door. A later review pulled together the processes behind it, among them the way we infer who we are from what we have already done, the pull to stay consistent with that, and a growing sense of commitment. The first yes changes how you see yourself, and the second yes follows from the new picture. The review also found the technique can fail or even backfire, which matters later, because it makes the drift a tendency that can be interrupted.
Now apply that to a working week. The first time you pick up a colleague's missed deadline, you are a helpful person doing a favour. By the fourth time, you are the person who picks up missed deadlines, both in your own eyes and in everyone else's. Researchers writing about organisational citizenship gave this the name job creep: the slow process by which extra effort, offered freely, becomes the expected standard. Another strand of work named its end state compulsory citizenship, the going-the-extra-mile that is no longer optional in anything but name.
The costs of that end state are well measured. A study of 245 employees found that feeling pressure to be a good organisational citizen did raise how much people helped, and it also went with more conflict between work and family, more conflict between work and leisure, more job stress and a greater intention to quit, even after accounting for workload and hours. A study of 388 supervisor and employee pairs found that compulsory citizenship actually reduced the voluntary kind over time, because it weakened how much people identified with the organisation. And two studies found something that should interest every executive reading this: when people felt compelled to go above and beyond, they came to feel entitled, and that entitlement was linked to more deviant behaviour afterwards, at work and outside it. Pressure to help, a further study found, worked through a rise in negative feelings, and it hurt least in people whose own reasons for helping were intrinsic.
So forced generosity is expensive for everyone. What about the voluntary kind?
Helping does feel good, and that deserves saying before the costs. A ten-day study of 82 employees found that helping on a given day went with more positive feelings that day. It also got in the way of people's sense of progress on their own work, which pulled their well-being down again. A three-week diary study found that responding to requests for help drained people's self-regulatory resources at an increasing rate, so that each extra request cost more than the last, while seeing the good the help did topped those resources back up. People with strong prosocial motives were drained more and replenished less. A ten-day study of 91 employees found that the depletion from helping made people less likely to help afterwards, and more likely to act in their own political interest. A time-lagged study in contact centres found that conscientious going-beyond was linked to more emotional exhaustion and work-family conflict, and most strongly in the people who were also doing their core jobs best.
And the career cost falls where you would least expect it. Archival data from 3,680 employees in a professional services firm found that, holding their core work constant, people who spent more time on citizenship behaviour received smaller pay rises and advanced more slowly. Where the extra effort was supposed to be rewarded, it was quietly taxed.
That tax is not charged equally. In two experiments, helping at work improved evaluations of men but not of women, and declining to help lowered evaluations of women but not of men; a third study found that helping at work was seen as less optional for women. A survey of 440 people found that helping the organisation was linked to more promotions, and more strongly for men than for women. If you have ever wondered why the generosity of the most capable women in an office seems to be banked rather than credited, this is part of the answer.
Then there is the finding I would put on the wall of every executive suite. In a set of six studies, people with high self-control, the dependable ones, were held to higher performance expectations and were assigned more work by observers. The observers also believed those dependable people were putting in less effort than less dependable colleagues, while the dependable people themselves reported no such thing. People high in self-control felt more burdened by the reliance of their coworkers and their partners, and in the romantic sample that burden went with lower satisfaction in the relationship. The more reliable you are, the more you are given, and the less anyone thinks it costs you.
The more reliable you are, the more work arrives, and the less anyone believes it is costing you, because you make it look like nothing.
I used to say that what you absorb silently becomes the new standard, and the research on whole workplaces bears that out. A nine-month study of a software team found a time famine kept going by constant interruptions, a crisis mentality and rewards for individual heroics; when the team changed how it used its time, collective productivity went up. And an ethnography of a university found managers widening workloads without formal authority through sequences the researchers called overtures and the interactional trebuchet, each designed to make the next ask hard to refuse.
Try this. Pick the three things in your week that began life as a favour. For each one, write down when it started, whether anybody ever agreed it was now yours, and what it displaced. You are not deciding yet whether to keep it. You are turning a drift into a decision, which is the only form in which it can be discussed.
The research behind this section
Freedman and Fraser (1966) found in two experiments that people who had agreed to a small request were more likely to comply with a larger one, including when a different person made the second request. Burger (1999) reviewed the foot-in-the-door research and several meta-analyses and identified self-perception, consistency, commitment, reactance, conformity and attribution processes that can each strengthen or weaken the effect, explaining both successful and unsuccessful demonstrations. Van Dyne and Ellis (2004) described job creep, the process by which citizenship behaviour comes to be expected as an overfulfilment of obligations. Vigoda-Gadot (2006) theorised compulsory citizenship behaviour as a dark side of the good soldier syndrome. Bolino, Turnley, Gilstrap and Suazo (2010) studied 245 employees and found that citizenship pressure was related to more citizenship behaviour and, controlling for it and for role overload and hours worked, to work-family conflict, work-leisure conflict, job stress and intentions to quit. Zhao, Peng and Chen (2014) studied 388 supervisor and subordinate pairs in 67 teams and found that compulsory citizenship behaviour reduced voluntary citizenship behaviour through impaired organisational identification, more so under low interactional justice. Yam, Klotz, He and Reynolds (2017) found in two multisource field studies and an experiment that citizenship behaviour performed under external compulsion led to psychological entitlement, which was linked to interpersonal and organisational deviance and to deviance outside the organisation. Koopman and colleagues (2020) found in two studies that helping pressure was indirectly related to deviant behaviour through negative affect, and that intrinsic motivation for citizenship weakened the effect. Bolino, Klotz, Turnley and Harvey (2013) reviewed the dark side of citizenship behaviour. Koopman, Lanaj and Scott (2016) surveyed 82 employees over ten workdays and found daily citizenship behaviour associated with positive affect and with interference in perceived work goal progress. Lanaj, Johnson and Wang (2016) used daily diaries over three work weeks and found that responding to help requests depleted regulatory resources at an increasing rate, that perceived prosocial impact replenished them, and that prosocially motivated employees were depleted more and replenished less. Gabriel, Koopman, Rosen and Johnson (2018) studied 91 employees over ten workdays and found helping depleting, making subsequent helping less likely and self-serving political behaviour more likely. Deery, Rayton, Walsh and Kinnie (2017) found in contact-centre employees that conscientiousness citizenship behaviour was associated with emotional exhaustion and work-family conflict, most strongly among high in-role performers. Klotz, Bolino, Song and Stornelli (2018) identified profiles of citizenship behaviour that predicted performance ratings, status and citizenship fatigue. Bergeron, Shipp, Rosen and Furst (2013) analysed archival data from 3,680 employees in a professional services firm and found that, controlling for time spent on task performance, employees who spent more time on citizenship behaviour had lower salary increases and advanced more slowly. Heilman and Chen (2005) found in two experiments that altruistic citizenship behaviour enhanced evaluations and recommendations of men but not women, and withholding it diminished evaluations of women but not men, with a third study finding work-related altruism seen as less optional for women. Allen (2006) surveyed 440 employees and found citizenship behaviour directed at the organisation related to promotions, more strongly for men than for women. Koval, vanDellen, Fitzsimons and Ranby (2015) found across six studies that observers held higher performance expectations for, and assigned greater workloads to, people with high self-control, judged them to expend less effort, and that people high in self-control reported greater burden from the reliance of coworkers and romantic partners, with lower relationship satisfaction. Kim and Glomb (2014) studied 4,874 university staff in 339 work groups and a further multisource sample and found high performers more likely to be targets of victimisation, through fellow group members' envy. Perlow (1999) studied a software engineering team for nine months and found a time famine perpetuated by interruptions, a crisis mentality and rewards for individual heroics, with collective productivity improving when the way time was used changed. Martin and Wilson (2024) found university managers seeking compliance with increased workloads through overtures and what they called the interactional trebuchet. Hassard and Morris (2022) studied middle managers across two phases, 2002 to 2006 and 2015 to 2019, and found the employment boundary stretched inside the workplace and breached outside it, attributing this to strategic and political choice as well as technology.
Holding the line, warmly
Most of the people I have worked with do not struggle to draw a boundary. They struggle with the feeling afterwards, the sense of having been too much, which is what sends them back to apologise and quietly undo it. The research on assertiveness has a great deal to say about that feeling, and most of it is reassuring.
Start with the shape of the thing. Across three studies, people seen as markedly low in assertiveness and people seen as markedly high were both judged less effective as leaders than people in the middle. Too much damaged relationships; too little meant goals were not reached. And perceptions of getting it wrong in one direction or the other were widespread. The target is the middle of the range.
Now the useful part. In a set of four studies of negotiation, people's views of their own assertiveness were only loosely linked to how their counterparts saw them. The people seen as pushing too hard or giving in too easily mostly believed they had got it about right. And many of the people their counterparts saw as having got it exactly right believed they had overstepped. The researchers called this the line-crossing illusion. Part of the cause, they found, was what they named strategic umbrage: counterparts putting on displays of discomfort, a sigh, a pause, a wounded look, which made the other person feel they had gone too far when they had not. And the illusion mattered, because it predicted how later rounds went.
If you have ever drawn a line with an executive and then spent the evening replaying their face, read that twice. The discomfort you saw may have been real, or it may have been the ordinary theatre of someone who would have preferred a different answer, and neither means you were wrong.
The moment you feel you have pushed too hard is, surprisingly often, the moment you got it about right.
The same group's review of assertiveness found that people assert themselves most effectively with precise proposals and with rationales, meaning an actual reason attached to the position. An older study offers the warmest version of that. When people rated different ways of standing up for oneself, plain assertion was seen as more appropriate and more effective than either giving in or being aggressive, though a little less polite and less pleasing to the other person than giving in. Adding what the researchers called extra consideration, an acknowledgement of the other person's position alongside the assertion, kept all of the effectiveness and made the assertion kinder, less hostile and more satisfying to the person on the receiving end.
Reasons matter in their own right. A meta-analysis of 54 samples found that explanations had strong effects on how fair a decision seemed and on whether people cooperated or retaliated afterwards, especially when the news was unwelcome. One detail surprised me: explanations that pointed to circumstances, which the researchers classed as excuses, did better than justifications arguing the decision was right. "The board papers need the whole of Thursday" lands better than "it would be inappropriate for me to take this on". And a much older field experiment found that for small requests, the shape of a reason did a lot of the work, while for more effortful ones people attended to what the reason actually said. For a boundary that matters, the reason has to be real.
There is a gendered cost here that the research is honest about. A meta-analysis of studies on reactions to dominant behaviour found that dominance hurt women's likeability more than men's, although the overall effect was small, and hurt their chances of being hired more. What decided it was the form. Explicit dominance, such as direct demands, dented women's likeability; implicit forms, such as steady eye contact, did not. And dominance made no difference to how competent women were judged to be, compared with men. A separate set of experiments found that the penalties successful women faced in likeability and perceived hostility were removed when there was clear evidence of their communal, caring side. The practical reading is uncomfortable and useful: a boundary delivered as a demand can cost you in liking, and the same boundary delivered as a plan, with the care visible, largely does not.
Now the disagreement, because you deserve both halves. A study of negotiations over price, published in 2019, found that people who wrote their offers in warm, friendly, polite language got worse deals than people who wrote them in tough, firm language, because firm offers drew better counteroffers. Their counterparts did not enjoy the tough negotiators any less. And most people believed the opposite would happen. So warmth is not free. My reading of the two literatures together is that warmth works when it is aimed at the person and does not soften the terms. "I know how much this launch matters to you, and I cannot take the travel on as well; here is who can" is warm about the person and firm about the line. "I'm so sorry, I would love to, it's only that I might possibly struggle this week" is warm about everything and firm about nothing, and the other person will hear it as an opening bid.
One last finding belongs here, because it covers the yes as well as the no. Across four studies, people who received help responded more warmly and were more inclined to reciprocate when they believed the helper had decided to help out of feeling for them, rather than because it was their role or because they had weighed up the costs. The damage from visible cost-benefit reasoning was greatest when the help was small. When you do say yes, say it like a person. It is the small, grudging, calculated yes that costs you most.
Try this. Before your next difficult conversation about scope, write the boundary as one sentence with three parts: something that shows you understand what matters to the other person, the line itself in plain words, and what you can offer instead or who else can help. Say it once. Then, when you feel the urge to soften it with a second sentence, notice the feeling, name it to yourself as the line-crossing illusion, and let the pause sit.
The research behind this section
Ames and Flynn (2007) found across three studies that people seen as either markedly low or markedly high in assertiveness were generally appraised as less effective leaders, linked to a trade-off between social outcomes (high assertiveness worsened relationships) and instrumental outcomes (low assertiveness limited goal achievement), and that perceptions of too much or too little assertiveness were widespread. Ames and Wazlawek (2014) found across four studies, including dyadic negotiations, that those seen as under- or over-assertive were likely to see themselves as appropriately assertive, that many people seen as appropriately assertive mistakenly believed they had been seen as over-assertive (the line crossing illusion), that counterparts' orchestrated displays of discomfort, termed strategic umbrage, were widespread and linked to the illusion, and that the illusion predicted outcomes in a multi-round negotiation. Ames, Lee and Wazlawek (2017) reviewed interpersonal assertiveness, its curvilinear consequences and ways of asserting effectively, including precise offers and rationales. Ames (2009) reviewed assertiveness and effectiveness in leadership and interpersonal dynamics. Woolfolk and Dever (1979) found that assertion was rated more appropriate and effective than nonassertion or aggression, less polite and less satisfying to the recipient than nonassertion, and that assertion plus extra consideration was comparable in effectiveness and appropriateness while being rated kinder, less hostile and more satisfying to recipients. Shaw, Wild and Colquitt (2003) meta-analysed 54 independent samples and found strong effects of explanations on justice judgements and on cooperation, retaliation and withdrawal, with excuses more beneficial than justifications and explanations more beneficial after unfavourable outcomes. Langer, Blank and Chanowitz (1978) found in three field experiments that unless a request called for an effortful response, people responded to the structure of a communication rather than its semantic content. Williams and Tiedens (2016) meta-analysed studies of evaluations of dominant behaviour and found that dominance hurt women's likeability relative to men's (d = −0.19, k = 63) and hireability (d = −0.58, k = 20), that explicit dominance affected women's likeability (d = −0.28) while implicit dominance did not (d = 0.03), and that dominance did not differentially affect perceived competence (d = 0.02, k = 31). Heilman and Okimoto (2007) found in three experiments that penalties in likeability, interpersonal hostility and boss desirability directed at successful female managers were mitigated when there was clear information that they were communal. Jeong, Minson, Yeomans and Gino (2019) found in a field experiment and a laboratory experiment that negotiators using a tough and firm style achieved better economic outcomes than those using a warm and friendly style, because firm offers elicited more favourable counteroffers, that counterparts did not report different satisfaction or enjoyment, and that lay beliefs predicted the opposite. Ames, Flynn and Weber (2004) found in four studies that recipients were more inclined to future interaction and reciprocation when they perceived help as based on affect rather than on role or cost-benefit calculation, with the negative impact of apparent cost-benefit thinking greatest when the help was small.
When a boundary slips
It will slip. There will be a week with a board meeting, a sick child and a crisis nobody saw coming, and you will do the thing you said you would stop doing, because in that moment it was the right call. The more useful question is what the slip does next.
Here is the finding that changed how I think about consistency. In a laboratory study and an experience-sampling field study, people treated fairly some of the time and unfairly at other times showed more physiological stress than people treated fairly all the time, and more than people treated unfairly all the time. Variability itself was the stressor, and it fed through to dissatisfaction and exhaustion. That research was about how supervisors treat staff, but I think it describes boundaries on both sides of the desk. A boundary that holds on Mondays and dissolves on Thursdays is harder on everybody, including the person on the other side, than one that is simply there. Being predictable is kinder than being lovely some of the time.
The most dangerous moment comes with the second request after you draw a line, rather than the first. A set of four studies found that people who had refused a request once were more likely to agree to a similar request afterwards, because turning someone down twice felt so much worse than turning them down once. The people asking did not realise this, and underestimated how often a second request would succeed. So if a boundary is going to go, it will probably go on the second ask, in a moment when you are tired of feeling unhelpful. Knowing that in advance is half the defence.
Consistency, in the research on leaders, has a name: behavioural integrity, the perceived alignment between what someone says and what they do, with particular attention to promises kept. It is a useful frame for your own boundaries, because what the other person learns is the pattern, not the statement. If you say evenings are for emergencies and then answer every evening message within ten minutes, the pattern is what they will believe.
But one slip is not a pattern, and the habit research is specific about this. In a study of 96 volunteers who each tried to build a new daily habit over twelve weeks, the time it took to become automatic ranged from 18 to 254 days, and missing a single opportunity did not materially affect the process. What mattered was coming back to it. A set of four experiments found that treating yourself with compassion after a failure, rather than either criticising yourself or trying to boost your self-esteem, made people more motivated to improve, more willing to make amends and more likely to put in extra effort after a setback. And a study of fresh starts found that people tackled their goals more readily after natural landmarks, the start of a week, a month or a new quarter, because those moments put past lapses into a previous chapter.
The research on broken deals between people and their employers is hopeful in the same way. A study of 109 young academics, tracked week by week after they reported a breach of their psychological contract, found that commitment could bounce back within ten weeks, that some people recovered far better than others, and that recovery went with how emotionally heavy the breach had been and how supported people felt afterwards. A study of 107 graduate newcomers over four waves, with 26 follow-up interviews, found a pattern the researcher called adaptive remediation: a breach triggered a dip and a pulling back, and then people acted, either fixing the breach directly or reframing the wider relationship, and the relationship recovered. The repair was something people did, not something that happened to them.
Slips cluster in the people who care most. A laboratory study and two field samples found that people prone to guilt put more effort into their work, which was what made them more committed. The same disposition that makes you excellent makes the unanswered message feel like a moral failure. It is worth knowing that about yourself, so that you can tell the difference between a real obligation and the ordinary discomfort of having a limit.
Try this. When a boundary slips, give it one sentence and a date. The sentence is to the other person, and it names the slip as an exception: "I picked this up this week because of the audit; from Monday it goes back to the finance team." The date is for you: the next Monday, or the first of the month, when the line resumes. Then do the kind thing for yourself that you would do for a colleague, and move on.
The research behind this section
Matta, Scott, Colquitt, Koopman and Passantino (2017) found in a laboratory study that variably fair treatment produced greater physiological stress than both consistently fair and consistently unfair treatment, and in a multilevel experience-sampling field study replicated the link between justice variability and stress, with daily stress mediating effects on job dissatisfaction and emotional exhaustion; supervisors with more self-control were less variable in their fair treatment. Newark, Flynn and Bohns (2014) found in four studies that help-seekers believed a previous refusal would make a potential helper less likely to agree to a subsequent similar request and underestimated actual compliance, and that potential helpers' willingness to comply stemmed from the discomfort of rejecting others twice. Simons (2002) set out behavioural integrity, the perceived alignment between a manager's words and deeds, with attention to promise keeping. Lally, van Jaarsveld, Potts and Wardle (2010) followed 96 volunteers building a daily habit over 12 weeks and found that the time to reach 95% of the asymptote of automaticity ranged from 18 to 254 days and that missing one opportunity did not materially affect habit formation. Breines and Chen (2012) found in four experiments that self-compassion after a weakness, transgression or failure increased motivation to improve, to make amends and avoid repeating a transgression, and time spent studying after an initial failure, compared with self-esteem and control conditions. Dai, Milkman and Riis (2014) found across three archival field studies that aspirational behaviours increased following temporal landmarks such as the start of a week, month or year. Solinger, Hofmans, Bal and Jansen (2016) tracked 109 young academics for up to ten weeks after a reported psychological contract breach and found that organisational commitment could recover, that recovery success related to the emotional impact of the breach and post-breach perceived organisational support, and that trajectories had a nonlinear component. Bankins (2015) studied 107 graduate newcomers over four waves and interviewed 26 and developed an adaptive remediation model in which breach triggered sensemaking and withdrawal, followed by coping actions that either repaired the breach directly or reappraised the wider relationship. Flynn and Schaumberg (2012) found in a laboratory study and two field samples that guilt-proneness was associated with higher affective organisational commitment, driven by greater task effort. Gouldner (1960) analysed the norm of reciprocity as a near-universal moral norm that helps sustain stable social systems.
The executive's side of the line
Most of this piece has been written to the person drawing the boundary. This section is for the person on the other side of it, and for the version of you who will one day be there.
I have worked with founders who did not realise how much they relied on me until they paused long enough to see it. One of them said it plainly: "I depend on you in ways I never intended. You make my pressure feel lighter." That is the whole dynamic in two sentences, and none of it is malicious. The research explains why a good leader can miss it.
Power changes what people notice. In a set of experiments, people put into a high-power frame of mind were less likely to take another person's visual perspective, more likely to assume others shared knowledge they alone held, and less accurate at reading other people's emotional expressions. A second set of six studies found that under power, people approached others more on the basis of how useful they were for the goal in hand. Put those together with the finding that observers credit dependable people with less effort than they are actually spending, and you have an executive who is not unkind, simply badly placed to see the load beside them. The person who makes the pressure feel lighter is, almost by definition, the person whose own pressure is hardest to see.
The executive also sets the size of the role. A longitudinal field experiment on changing how broadly people saw their jobs found that interventions led by supervisors had the stronger influence early, when expectations were still unsettled, and that peers mattered more once the expectations became clearer. The first months with a new executive are when the edges are drawn, and the executive's pen is the heavier one.
What executives do when asked for support matters more than they may realise. Two studies, one a field study of 280 employees and one an experiment with 267 participants, found that asking a supervisor for support and not receiving it lowered people's sense that their basic needs at work were being met, and through that harmed their well-being, performance and relationships. The unanswered request cost something in its own right. A meta-analysis of 558 studies of perceived organisational support, the sense that the organisation values your contribution and cares about your well-being, found that leadership was among the things that built it, and that it fed through to performance and well-being.
And the way work is handed over shapes whether it is welcomed. A set of studies on delegation found that people were more likely to resist delegated work when the task was undesirable and offered no development, and when it was handed over rudely, because both weakened their commitment to the person delegating. The size of the ask mattered less than what it was and how it arrived.
So here is what I would ask an executive to do, if they want the boundaries of the person beside them to hold. Ask, at least twice a year, what has crept into the role that nobody agreed. Treat a boundary as information about capacity rather than a verdict on commitment. When you add something, name what comes off. Answer requests for support, even when the answer is not yet. And notice the second ask: if the person beside you has already said no once, asking again is not neutral, because you are asking someone who finds refusing twice harder than you do.
The recruitment agency study holds up a final mirror. Executive partners give exactly that kind of emotional help upwards, and it is very often received as simply part of the role. As one senior leader admitted to me once, "I never realised how much I expected from you until I stopped to consider what it would feel like to carry me." That is the conversation worth having, and it goes better when it is had on purpose, on a calm Tuesday, than in the middle of a crisis.
Try this. If you have a good relationship with your executive, share the three columns from the first section with them and ask them one question: "Which of these would surprise you?" If you are the executive, ask the person who supports you the same question in reverse. The answer is where the next conversation about boundaries should start.
The research behind this section
Galinsky, Magee, Inesi and Gruenfeld (2006) found across four experiments and a correlational study that power was associated with a reduced tendency to adopt another person's visual perspective, to take into account that others lacked one's privileged knowledge, and to judge others' emotional expressions accurately. Gruenfeld, Inesi, Magee and Galinsky (2008) found across six studies that under conditions of power, approach towards a social target was driven more by the target's usefulness for the perceiver's goals. Parke, Tangirala and Hussain (2021) used a longitudinal quasi-field experiment and a qualitative follow-up and found that supervisor-led role interventions had stronger effects on citizenship behaviour early in the change process, while peer-led interventions had positive and synergistic effects later, as expectations about behaviour became settled. McIlroy, Parker and McKimmie (2021) found in a field study of 280 employees and an experiment with 267 participants that unanswered requests for supervisor support indirectly harmed well-being, performance and relational outcomes through lower need satisfaction. Kurtessis and colleagues (2017) meta-analysed 558 studies and found that organisational support theory successfully predicted antecedents of perceived organisational support, including leadership, and its consequences for employee orientation, performance and well-being. Park, Carter and Phillips (2023) found in two experiments and a critical incident study that delegation of non-developmental and undesirable tasks, and supervisor incivility during delegation, increased subordinates' resistance to delegation through weaker commitment to the supervisor. Koval, vanDellen, Fitzsimons and Ranby (2015), described in the earlier section, found that observers judged people with high self-control to expend less effort. Toegel, Kilduff and Anand (2013), also described earlier, found that emotional help given by managers was seen by them as discretionary and by employees as part of the role.
The boundary check
Every idea in this piece comes down to one moment: there is something in your role that arrived as a favour and has stayed, and you need to decide whether to keep it, renegotiate it or hand it back. So here is a tool for that moment.
Put in the thing that has crept in, and answer eight questions about how it arrived, who thinks it is yours, what it costs and how the conversation about it would go. It will tell you which of the four lines is weakest, what to say and when, and what to watch for afterwards, and it gives you a note for the conversation to copy.
Where to start
If you take one thing from all of this, make it the three columns. Most boundaries that fail were never agreed in the first place; they were assumed by one person and invisible to the other. The research on role definitions is consistent across countries, jobs and decades: the people doing the work and the people directing it draw the edges of the job in different places, and the wider line usually belongs to the person with the power. Writing down what you own, what you support and what is not yours turns that invisible difference into something two people can talk about.
After that, there are three moves worth making.
When you negotiate something new, negotiate what goes in the same breath, because the research found the growth deals were the ones that came with more unpaid overtime. When you hold a line, say it once, with care for the person and firmness on the terms, and treat the feeling of having overstepped as the line-crossing illusion it probably is. And when it slips, name the slip as an exception and give the line a date to resume; one lapse does not undo a habit, but drifting quietly does.
I have put the three pages I would use into a download: a role map with the three columns and a space to compare versions, a one-page working agreement with a review date, and a slip log for the weeks when the line moves, so that you can see whether it is an exception or a pattern.
Boundaries have a reputation for being about keeping people out. In a close working partnership they turn out to be much more like good architecture: the walls are what let everyone know where the rooms are, and the doors are what make the whole building usable. The partnerships that last are rarely the ones with no lines. They are the ones where both people know where the lines are, and have agreed them.
The part of this that shows up in how you are seen at work sits inside what the Connected Leader Assessment calls Sustainable Performance: how you manage your capacity, how clearly the scope of your role is agreed, and whether the standard you set is one you can keep up for years rather than weeks. It is the dimension that decides whether your best work is repeatable or only heroic.
The assessment is free, takes about twelve minutes and gives you a read across all five dimensions. If you want to go further on this one, the Skill Sprints are short, focused programmes on a single dimension, and the Irreplaceable Audit is the place to separate the work only you can do from the work that simply drifted onto your desk, which is the first line any boundary worth keeping has to draw.
Stay savvy.
Meg ✌️
Free tools
The boundary check, the interactive tool in this post, free
The Boundary Pages, the download in this post, free
The Connected Leader Assessment, twelve minutes, free, with a read on Sustainable Performance, the dimension this post sits in
The Irreplaceable Audit, forty-eight activities across four kinds of work, free, for drawing the edges of your role before anyone else does
P.S. Want to Go Deeper on Boundaries That Stick?
If this resonated and you want to go further, here is the evidence the piece is built on. These are the papers that shaped my thinking on what makes a line hold once it is drawn in a close working partnership, and they are well worth your time.
Every DOI below has been checked. Where I could not confirm a figure from the paper itself, I have left the figure out rather than guess.
📃 Koval and colleagues (2015) Six studies. Dependable people were given more work and credited with less effort. https://doi.org/10.1037/pspi0000015
📃 Ames and Flynn (2007) Three studies. Too little and too much assertiveness both made leaders look less effective. https://doi.org/10.1037/0022-3514.92.2.307
📃 Ames and Wazlawek (2014) The line-crossing illusion: people who got it right often believed they had pushed too hard. https://doi.org/10.1177/0146167214525474
📃 Ames, Lee and Wazlawek (2017) Review of interpersonal assertiveness, including precise proposals and rationales. https://doi.org/10.1111/spc3.12317
📃 Ames (2009) Review: assertiveness helps up to a point, and then it costs. https://doi.org/10.1016/j.riob.2009.06.010
📃 Morrison (1994) 317 clerical workers. People doing the same work disagreed about which parts were the job and which were extra. https://doi.org/10.2307/256798
📃 Lam, Hui and Law (1999) 431 supervisor and employee pairs in four countries. Supervisors drew the job wider than the people doing it. https://doi.org/10.1037/0021-9010.84.4.594
📃 Van Dyne and Ellis (2004) Job creep: how extra effort offered freely comes to be expected. https://doi.org/10.1093/oso/9780199269136.003.0009
📃 Vigoda-Gadot (2006) Compulsory citizenship: the dark side of the good soldier. https://doi.org/10.1111/j.1468-5914.2006.00297.x
📃 Rousseau (1989) The founding statement of psychological and implied contracts at work. https://doi.org/10.1007/bf01384942
📃 Robinson and Rousseau (1994) 128 graduates. 54.8% said the deal had been broken within two years, with lower trust and more intention to leave. https://doi.org/10.1002/job.4030150306
📃 Zhao and colleagues (2007) Meta-analysis. A broken deal hit nearly every outcome, through feelings of violation and mistrust. https://doi.org/10.1111/j.1744-6570.2007.00087.x
📃 Solinger and colleagues (2016) 109 academics. Commitment could bounce back within weeks of a broken deal, helped by support. https://doi.org/10.1002/job.2047
📃 Hornung, Rousseau and Glaser (2008) 887 employees. Flexibility deals cut unpaid overtime; development deals came with more of it. https://doi.org/10.1037/0021-9010.93.3.655
📃 Liao, Wayne and Rousseau (2016) Review and meta-analysis of 23 studies of personally negotiated work arrangements. https://doi.org/10.1002/job.1959
📃 Sin, Nahrgang and Morgeson (2009) 64 samples, 10,884 pairs. Leader and follower agreed only moderately, and more with time and contact. https://doi.org/10.1037/a0014827
📃 Galinsky and colleagues (2006) Power reduced people's tendency to take others' perspectives and read their emotions. https://doi.org/10.1111/j.1467-9280.2006.01824.x
📃 Kim and Glomb (2014) 4,874 staff. High performers were more often targets of mistreatment, through envy. https://doi.org/10.1037/a0035789
📃 Lanaj, Johnson and Wang (2016) Three weeks of diaries. Each help request drained people more than the last; seeing the impact topped them up. https://doi.org/10.1037/apl0000118
📃 Koopman, Lanaj and Scott (2016) 82 employees over ten days. Helping lifted mood and got in the way of progress on one's own work. https://doi.org/10.5465/amj.2014.0262
📃 Gabriel and colleagues (2018) 91 employees over ten days. Helping depleted people, who then helped less and politicked more. https://doi.org/10.1111/peps.12229
📃 Koopman and colleagues (2020) Pressure to help raised negative feelings and deviance, less so for intrinsically motivated helpers. https://doi.org/10.1111/peps.12354
📃 Bettenhausen and Murnighan (1985) 19 new groups. Norms formed from members' scripts and were either tacitly accepted or openly challenged. https://doi.org/10.2307/2392667
📃 Lally and colleagues (2010) 96 volunteers. Habits took 18 to 254 days, and missing one day did not derail them. https://doi.org/10.1002/ejsp.674
📃 Rizzo, House and Lirtzman (1970) The original measures of role conflict and role ambiguity, and their links to satisfaction, anxiety and leaving. https://doi.org/10.2307/2391486
📃 Kurtessis and colleagues (2017) 558 studies. Feeling valued by the organisation fed performance and well-being. https://doi.org/10.1177/0149206315575554
📃 Bolino and colleagues (2013) Review of the dark side of organisational citizenship. https://doi.org/10.1002/job.1847
📃 Heilman and Chen (2005) Helping lifted ratings of men, not women; not helping lowered ratings of women, not men. https://doi.org/10.1037/0021-9010.90.3.431
📃 Williams and Tiedens (2016) Meta-analysis. Explicit dominance cost women likeability; implicit dominance did not, and competence was unaffected. https://doi.org/10.1037/bul0000039
📃 Jeong and colleagues (2019) Firm offers drew better counteroffers than warm ones, with no loss of liking. https://doi.org/10.1287/mnsc.2018.3199
📃 Shaw, Wild and Colquitt (2003) 54 samples. Explanations improved fairness and cooperation; excuses beat justifications. https://doi.org/10.1037/0021-9010.88.3.444
📃 Matta and colleagues (2017) Being treated fairly only some of the time was more stressful than consistent unfairness. https://doi.org/10.5465/amj.2014.0455
📃 Dabos and Rousseau (2004) 80 pairs. Where both sides' versions of the deal matched, productivity and careers did better. https://doi.org/10.1037/0021-9010.89.1.52
📃 Mathieu and Rapp (2009) 32 teams. A good charter plus a good strategy produced the highest sustained performance. https://doi.org/10.1037/a0013257
📃 Truss and colleagues (2013) 1,011 secretaries. Better qualified and paid, a minority doing managerial work, prospects still bleak. https://doi.org/10.1111/j.1468-0432.2012.00587.x
📃 Star and Strauss (1999) The classic account of visible and invisible work. https://doi.org/10.1023/a:1008651105359
📃 Breines and Chen (2012) Four experiments. Self-compassion after a failure raised the motivation to improve. https://doi.org/10.1177/0146167212445599
📃 Flynn and Schaumberg (2012) Guilt-prone people put in more effort, which made them more committed. https://doi.org/10.1037/a0024166
📃 Woolfolk and Dever (1979) Assertion with extra consideration was as effective as plain assertion, and kinder. https://doi.org/10.1016/s0005-7894(79)80028-3
📃 Bergeron and colleagues (2013) 3,680 employees. More time on extra effort meant smaller pay rises and slower advancement. https://doi.org/10.1177/0149206311407508
📃 Allen (2006) 440 people. Helping the organisation went with promotions, more strongly for men. https://doi.org/10.1111/j.0021-9029.2006.00006.x
📃 Earnest, Allen and Landis (2011) 52 studies, about 17,000 people. Realistic job previews worked mainly by signalling honesty. https://doi.org/10.1111/j.1744-6570.2011.01230.x
📃 Hatton (2017) The cultural, legal and spatial mechanisms that make work invisible and undervalued. https://doi.org/10.1177/0950017016674894
📃 Eatough and colleagues (2011) 42 studies. Ambiguity and conflict went with less helping; overload went with a little more. https://doi.org/10.1037/a0021887
📃 Newark, Flynn and Bohns (2014) People who refused once found it harder to refuse again, and askers did not realise it. https://doi.org/10.1177/1948550613490967
📃 Burger (1999) Review of the foot in the door and the processes that make it work, or backfire. https://doi.org/10.1207/s15327957pspr0304_2
📃 Freedman and Fraser (1966) The foot in the door: agreeing to a small request made a larger one more likely, even from someone else. https://doi.org/10.1037/h0023552
📃 Langer, Blank and Chanowitz (1978) For small requests the shape of a reason did the work; for bigger ones the content mattered. https://doi.org/10.1037/0022-3514.36.6.635
📃 Lai, Rousseau and Chang (2009) 20 research groups. Colleagues accepted a personal deal more readily from friends, and when they expected their own turn. https://doi.org/10.1037/a0013506
📃 Gilboa and colleagues (2008) 169 samples, 35,265 employees. Role ambiguity was among the stressors most strongly tied to poorer performance. https://doi.org/10.1111/j.1744-6570.2008.00113.x
📃 Ashford and Black (1996) Newcomers negotiated changes to their jobs as one of several ways of gaining control. https://doi.org/10.1037/0021-9010.81.2.199
📃 Rousseau (2001) Schemas, promises and mutuality as the building blocks of the unwritten deal. https://doi.org/10.1348/096317901167505
📃 Gouldner (1960) The norm of reciprocity as a near-universal moral rule. https://doi.org/10.2307/2092623
📃 Deery and colleagues (2017) Contact-centre staff. Conscientious extra effort went with exhaustion, most for the best performers. https://doi.org/10.1002/hrm.21815
📃 Yam and colleagues (2017) Going the extra mile under compulsion bred entitlement, and entitlement bred deviance. https://doi.org/10.5465/amj.2014.0234
📃 Klotz and colleagues (2018) Profiles of citizenship behaviour, and their links to ratings, status and citizenship fatigue. https://doi.org/10.1002/job.2259
📃 Park, Carter and Phillips (2023) Undesirable, non-developmental tasks and rude delegation raised resistance to delegated work. https://doi.org/10.1007/s10869-023-09898-4
📃 Parke, Tangirala and Hussain (2021) Supervisors shaped how broadly people saw their roles most strongly early on. https://doi.org/10.1037/apl0000848
📃 Kang and Ryan (2016) 469 workers. Helping was more likely to be counted as part of the job for senior people. https://doi.org/10.1186/s40064-016-3097-1
📃 Turnipseed and Rassuli (2005) Managers and employees saw going-the-extra-mile differently; the authors urged managers to say what is extra and what is expected. https://doi.org/10.1111/j.1467-8551.2005.00456.x
📃 Organ (1997) The field's founder conceded that 'extra' effort was rarely extra, beyond the job or unrewarded in practice. https://doi.org/10.1207/s15327043hup1002_2
📃 McIlroy, Parker and McKimmie (2021) 280 employees and 267 participants. Asking for support and not getting it did its own harm. https://doi.org/10.1037/ocp0000270
📃 Martin and Wilson (2024) Managers widened workloads without authority through overtures and the 'interactional trebuchet'. https://doi.org/10.1177/08912416241265704
📃 Hassard and Morris (2022) Middle managers over fifteen years. Work boundaries were stretched inside work and breached outside it. https://doi.org/10.1177/00187267211003123
📃 Simons (2002) Behavioural integrity: the fit between what a manager says and does. https://doi.org/10.1287/orsc.13.1.18.543
📃 Conway and Briner (2002) Daily diaries. Broken and exceeded promises turned up regularly, in almost any part of work. https://doi.org/10.1002/job.139
📃 Toegel, Kilduff and Anand (2013) Managers saw emotional help as a favour; staff saw it as the job. The managers ended up disappointed. https://doi.org/10.5465/amj.2010.0512
📃 Gruenfeld and colleagues (2008) Six studies. Under power, people approached others according to how useful they were. https://doi.org/10.1037/0022-3514.95.1.111
📃 Dai, Milkman and Riis (2014) Fresh starts: people tackled goals more after landmarks such as a new week or month. https://doi.org/10.1287/mnsc.2014.1901
📃 Bankins (2015) 107 graduates. People repaired broken deals through their own coping actions. https://doi.org/10.1002/job.2007
📃 Zhao, Peng and Chen (2014) 388 pairs. Compulsory helping reduced the voluntary kind by weakening identification. https://doi.org/10.1080/00223980.2013.768591
📃 Heilman and Okimoto (2007) Penalties for successful women disappeared when their communal side was clear. https://doi.org/10.1037/0021-9010.92.1.81
📃 Ames, Flynn and Weber (2004) Help seen as coming from feeling won more goodwill than help seen as role or calculation. https://doi.org/10.1177/0146167203261890
📃 Lavee and Kaplan (2022) 185 interviews. Informal extra work became invisible to the organisation, society and the workers themselves. https://doi.org/10.1111/gwao.12839
📃 Bolino and colleagues (2010) 245 employees. Pressure to go the extra mile went with stress, work-family conflict and intention to quit. https://doi.org/10.1002/job.635
📃 Liu and colleagues (2013) 230 employees, 102 supervisors. Personal deals raised proactivity and commitment through fairness and feeling valued. https://doi.org/10.1037/a0032571
📃 Perlow (1999) Nine months with a software team. Interruptions and rewarded heroics kept a time famine going. https://doi.org/10.2307/2667031
📃 Hornung, Rousseau and Glaser (2009) 263 supervisors. Workload reductions were granted mostly where the organisation had already fallen short. https://doi.org/10.1108/02683940910996770
P.P.S. Why Do I Even Have the Nerve to Write This?
Fair question.
I am someone who has spent years working alongside remarkable executives, navigating chaos, translating vision into structure and figuring things out, sometimes beautifully and sometimes the very hard way. I have had the privilege of being the steady right hand to leaders who move fast, take risks and expect a lot. That environment has shaped me more than any classroom ever could.
I also happen to have an MBA, and I am studying psychology because people fascinate me. How we work. Why we disconnect. What actually holds us together when the pressure rises.
But really, none of that is the point.
I am here because I have lived the erosion and rebuilt from it. I have seen the cost of carrying too much and the relief that arrives when you finally stop disappearing inside your own competence.
So I share what I have learned in case it helps someone else. Take what is useful. Leave the rest. And remember this one truth that professionals like you often forget:
You are probably doing far better than you give yourself credit for.
Comments