.png)
Politics, Played Well: What the Research Says About Influence, Informal Networks and Using Your Position Ethically.
Updated: Sep 28
The proposal went through the meeting beautifully. Everyone nodded, two people said it was overdue, and the chair asked for a paper by the end of the month. Then somewhere between the meeting room and the lifts it quietly died, and nobody could quite say who had killed it. The person who could have told you was the one who had booked the room, sent the papers and heard the conversation by the coffee machine afterwards, and in a great many organisations that person is an Executive Assistant or a Chief of Staff.
I remember the first time I realised how political the workplace could be. I was not involved; I was watching from the sidelines as my leaders worked their way through shifting alliances and unspoken rules, and I saw brilliant ideas stall for want of the right backing while people who understood influence moved ahead of people who were simply very good. I wrote about it early last year with a study of British managers at its heart, and the research around that study turns out to be far richer, more practical and more hopeful than a single survey could show.
What you will get out of this
What 250 managers said about office politics when asked directly, and the four quite different ways people make sense of it
What a political atmosphere costs the people working in it, measured across more than 44,000 people, and the two things that soften it
The four parts of political skill, which of them predicts performance, and why the answer to "does it help?" depends on the room
Why knowing who asks whom for advice is itself a source of influence, and who tends to get that map wrong
What sitting at the hub of information does to you, and the difference between the broker who joins people up and the broker who plays them off
Where the line sits between self-serving politics and the kind that helps everyone, including an uncomfortable finding about charm
How ideas actually get through a political organisation, from studies of 68 change initiatives and more than a hundred strategic decisions
The political read: a tool that looks at something you are trying to move and tells you where to start
The Politics Pages: three pages for mapping the informal organisation, building support for an idea and keeping yourself honest
What office politics actually is
The most useful thing to know about office politics is how ordinary it is. In 2008 a researcher surveyed 250 British managers about their experience of organisational politics, and political behaviour appeared to be common. Most of them viewed it as ethical and necessary, and they credited political tactics with parts of organisational effectiveness, change, resourcing and reputation. The tactics they met most often were hardly the stuff of television drama: networking, using key players to support initiatives, making friends with the people who hold power, bending the rules and a certain amount of self-promotion. The darker tactics, misinformation, spreading rumours and keeping what the survey called "dirt files", were reported as rare, although not unknown.
Two further findings from that survey are worth holding onto. Eighty per cent of those managers had no training in this area at all, which means most people are learning it the way I did, by watching. And the pattern of their answers about reciprocity and the need to act ruthlessly when faced with political behaviour suggested an attitude the author summed up in his title: “you stab my back, I'll stab yours”. Politics was accepted, widely practised and rarely taught.
Researchers have spent forty years trying to pin the subject down. One early theorist described the organisation as a political arena and argued that politics sometimes captures an organisation in whole or in part, from a contained confrontation to a place where conflict is moderate but everywhere. The same paper went on to describe the functional roles politics plays, which is a polite way of saying that it sometimes gets things done that the formal system cannot.
Most of the later research splits the subject in three. There is political behaviour, meaning what people actually do. There are perceptions of organisational politics, meaning how political a place feels to the people in it. And there is political skill, meaning how well a person reads and influences the people around them. A 2019 review argued that even that division is too narrow, and the most recent review of the perceptions research spends much of its length separating politics from neighbouring ideas. That separation is real: a study of 418 employees found that perceived politics, perceived fairness and perceived support from the organisation were distinct from one another, which matters because it means a place can feel fair and still feel political.
What makes a place feel political? An early pair of studies across four organisations found that feedback, job autonomy, skill variety and the chance of promotion all helped explain how political people thought their workplace was. People who knew how they were doing, had room to act and could see a route upwards saw less politics around them.
The most human research on the subject is a set of in-depth interviews with fourteen employees across three organisations. The researchers found that people made sense of politics at four distinct levels. At the reactive level politics was destructive and manipulative. At the reluctant level it was a necessary evil. At the strategic level it was a useful way of getting things done. And at the integrated level it was simply central to how the organisation functioned and made decisions. The same people described political behaviour in terms of five familiar kinds of power, among them connection power, information power and personal power, and they described both good and bad outcomes.
Two further studies explain why those four levels can sit side by side in one office. A study of 119 people found that positive and negative politics were separate perceptions, not two ends of a single line, and that seeing positive politics added to the prediction of how satisfied people were with their job, their supervisor and their colleagues, over and above seeing negative politics. Another found that people rated moderate amounts of negative politics more favourably than a great deal or none at all, that politics with nearby impact was judged more harshly as it became more frequent, and that positive politics was rated more favourably the more of it there was. People are not against politics. They are against politics that lands on them.
Office politics is mostly the informal half of how decisions get made, and the only real choice is whether you understand it.
Try this. Think of the last time something at work went a way you did not expect, and write one sentence about why. Then look at the sentence. If it describes a villain, you were reading it at the reactive level. If it describes how the decision was actually made, who was consulted, who was not, who talked to whom beforehand, you were reading it at the strategic level, and you have something you can use next time.
The research behind this section
Buchanan (2008) surveyed 250 British managers and found that political behaviour appeared to be common, that most viewed it as ethical and necessary, that aspects of organisational effectiveness, change, resourcing and reputation were attributed to political tactics, and that 80% had no training in this area; frequently experienced tactics included networking, using key players to support initiatives, making friends with power brokers, bending the rules and self-promotion, while misinformation, spreading rumours and keeping "dirt files" were experienced as rare but not unknown, and responses about reciprocity and the need to act ruthlessly implied an attitude of "you stab my back, I'll stab yours". Mintzberg (1985) set out four types of political arena, from the complete arena to the politicised organisation, and discussed the functional roles of politics in organisations. Ferris and colleagues (2002) reviewed theory and research on perceptions of organisational politics and proposed future directions. Ferris and colleagues (2019) argued that research divided into perceptions of politics, political behaviour and political skill had been artificially constrained, and proposed a framework of political characteristics, actions and outcomes. Hochwarter and colleagues (2020) reviewed three decades of perceptions of politics research, differentiated it from political behaviour and overlapping constructs, and summarised the meta-analytic evidence. Kacmar and Carlson (1997) evaluated and refined the Perceptions of Politics Scale across nine samples totalling 2,758 respondents. Andrews and Kacmar (2001) found in 418 full-time employees that perceptions of politics, organisational support, and procedural and distributive justice showed discriminant validity. Ferris and Kacmar (1992) found across four organisations that feedback, job autonomy, skill variety and opportunity for promotion contributed significantly to explaining perceptions of organisational politics. Landells and Albrecht (2017) interviewed 14 employees across three organisations and interpreted their perceptions of politics at four levels, reactive, reluctant, strategic and integrated, with political behaviours described in terms of connection, information, coercive, positional and personal power, and with both positive and negative outcomes. Fedor and colleagues (2008) found in 119 respondents that positive and negative politics were separate perceptions, and that positive politics perceptions contributed significantly beyond negative ones to predicting satisfaction with the job, supervision and co-workers and fulfilment of the psychological contract. Maslyn, Farmer and Bettenhausen (2017) found in two samples that moderate levels of negative politics were evaluated more favourably than high or low levels, that politics with nearby impact yielded more negative evaluations as frequency increased, and that positive politics was evaluated more favourably the more it occurred.
What a political atmosphere costs, and what softens it
So what does working in a place that feels highly political do to people?
Three meta-analyses have tried to answer that, and together they cover a great many people. The largest pulled together 118 samples and 44,560 participants. Perceived politics was most strongly related to lower trust in the organisation and lower interpersonal fairness, and it went with more stress, more burnout, stronger intentions to leave, more counterproductive behaviour, lower job satisfaction, less citizenship and lower job performance.
A 2009 meta-analysis put numbers on the main links, as correlations where zero means no relationship and one means a perfect one. Perceived politics correlated .48 with strain, .43 with intentions to quit, −.57 with job satisfaction and −.54 with commitment to the organisation. The links with behaviour were smaller: −.20 with task performance, and −.16 and −.20 with helping individuals and helping the organisation. The researchers found that politics reached performance through two routes, how people felt about the job and how strained they were, and they concluded that perceived politics behaves like a hindrance stressor, the kind of pressure that gets in the way rather than stretching you.
Here the good studies disagree, and it is worth seeing both halves. A 2008 meta-analysis of 79 samples and 25,059 people found the same strong links with satisfaction, commitment, stress and intentions to leave, and no significant relationship at all between perceived politics and in-role job performance, with age, sector and culture changing some of the links. The 2013 meta-analysis found a negative relationship with performance. Several single studies suggest why the answer keeps moving. A study of 303 public sector employees found that people responded to politics with neglect more than with leaving, and found only a weak negative link with supervisor-rated performance. A study of 577 people in 133 work units followed over six months found that perceived politics was linked to worse performance in flexible, loosely structured units and to *better* performance in tightly structured ones. A study of 177 financial services employees found an inverted U: a little politics went with higher satisfaction, and more loyal employees showed the same curve for performance, until the politics became extreme. And a 2025 study found that people who approach challenges treated a political atmosphere as a reason to build their relationship with their manager, and did better for it, while people who tend to avoid threats withdrew from their manager and did worse.
My reading is that politics reliably makes people less happy and more strained, and that whether it also makes them worse at their jobs depends on the structure around them and what they do in response. That second part is where you come in.
The feelings side has its own darker edge. Three samples of employees in private, public and voluntary organisations found that distress was an immediate response to workplace politics, and that the distress helped explain aggressive behaviour by those same employees afterwards. Politics does not stay politely inside the people who feel it.
What softens it? The research points to two things, and both are within an assistant's reach.
Information. A study of 150 employee and supervisor pairs found that workplaces with a strong informal feedback culture, where people routinely heard from supervisors and colleagues how they were doing, had lower perceptions of politics, and through that higher morale and better performance. The authors put it plainly: when people have better access to information about what is acceptable and wanted at work, they see less politics.
Understanding. The original model of perceived politics predicted that understanding how decisions are made would take some of the sting out, and a 1996 study supported it, finding that understanding and a sense of control softened the effects. A larger study of 786 government employees and 469 utility workers was more cautious: understanding softened the link between politics and job satisfaction, and did nothing for intentions to leave or anxiety. And a study of 231 employees added a twist. Psychological capital, a blend of confidence, hope, optimism and resilience, weakened the link between politics and both satisfaction and performance, and *strengthened* the link between politics and intentions to leave. The most resourceful people did better while they stayed, and were readier to go.
The antidote the research keeps finding to a political atmosphere is information, and assistants hold more of it than almost anyone.
Try this. Pick one decision coming up in the next month that people around you will be affected by, and find out, then share, three plain facts: who is deciding, when, and what they will be weighing. You are not leaking anything. You are giving people the understanding and the feedback that the research says makes a place feel less political, and you will be surprised how rarely anyone does it.
The research behind this section
Bedi and Schat (2013) meta-analysed 118 independent samples totalling 44,560 participants and found that perceptions of organisational politics were most strongly related to organisational trust and interactional justice, positively related to stress, burnout, turnover intentions and counterproductive work behaviour, and negatively related to job satisfaction, citizenship behaviour and job performance, with some effects moderated by publication status, geographical origin and measure. Chang, Rosen and Levy (2009) meta-analysed the relationship between perceived politics and outcomes and found correlations with strain (.48), turnover intentions (.43), job satisfaction (−.57), affective commitment (−.54), task performance (−.20), and citizenship towards individuals (−.16) and the organisation (−.20); attitudes mediated the effects on turnover intentions, attitudes and strain mediated the effects on performance, and perceived politics behaved as a hindrance stressor. Miller, Rutherford and Kolodinsky (2008) meta-analysed 79 samples from 59 studies involving 25,059 participants and found strong negative relationships with job satisfaction and organisational commitment, moderately positive relationships with job stress and turnover intentions, and a non-significant relationship with in-role job performance, with age, work setting and culture moderating some relationships. Vigoda (2000) found in 303 public sector employees in Israel that perceived politics was negatively related to job attitudes, positively related to intention to leave and more strongly to negligent behaviour, and weakly negatively related to supervisor-reported performance. Vigoda (2002) found in three samples of 155, 184 and 201 Israeli employees from the private, public and third sectors that job distress was an immediate response to organisational politics and a possible mediator between politics and aggressive behaviour. Yang (2017) followed 577 individuals and their supervisors in 133 work units over six months and found that perceived politics was negatively related to job performance in organic structures and positively related in mechanistic structures. Ellen and colleagues (2022) found in 177 financial services employees an inverted U-shaped relationship between perceived politics and job satisfaction, and inverted U-shaped relationships with both satisfaction and performance for more loyal but not less loyal employees. Fu and Peng (2025) found that approach-oriented employees appraised perceived politics as a challenge and responded by building relationships with supervisors, raising leader-member exchange and job performance, while avoidance-oriented employees appraised it as a threat, avoided interaction with supervisors and performed worse. Rosen, Levy and Hall (2006) found in 150 subordinate-supervisor dyads that higher quality feedback environments were associated with lower perceptions of politics, and that morale mediated the relationships between politics and work performance. Ferris and colleagues (1996) tested the perceptions of politics model and found support for most of its linkages, including the moderating effects of control and understanding. Kacmar and colleagues (1999) tested the model on 786 state government employees and 469 electric utility employees and found that understanding moderated the relationship between politics and job satisfaction but not intent to turnover or job anxiety. Abbas and colleagues (2014) found in 231 paired responses that psychological capital weakened the negative relationship of perceived politics with job satisfaction and supervisor-rated performance, but strengthened the relationship between politics and turnover intentions.
Political skill, in four parts
If politics is the weather, political skill is how well you dress for it. The researchers who built the main measure define it as the ability to understand others at work and to use that understanding to influence them in ways that serve personal or organisational goals, and they break it into four parts.
Social astuteness is reading people and situations accurately: noticing who is uneasy, what a silence means, why a meeting went the way it did.
Interpersonal influence is the ability to adapt how you come across so that people are persuaded, and often without them feeling pushed.
Networking ability is building and keeping relationships with a wide range of people, and knowing whom to call.
Apparent sincerity is coming across as honest and without an ulterior motive. It is the part that makes the other three work, because influence that looks like manipulation stops working the moment it is spotted.
The measure was developed across three investigations and seven samples. Political skill related to self-monitoring, political savvy and emotional intelligence, related negatively to anxiety, and was not correlated with general intelligence at all, which is a pleasing finding for anyone who has watched a very clever person make a mess of a room. It also predicted how managers' performance was rated in two samples.
A 2015 meta-analysis gathered the evidence together. Political skill went with higher self-efficacy, job satisfaction, commitment, productivity, citizenship, career success and personal reputation, and with lower physiological strain. It predicted task performance after taking account of both the big five personality traits and general mental ability. Part of the effect ran through reputation and self-belief. And when the four parts were tested separately, networking ability, interpersonal influence and apparent sincerity each predicted task performance, and social astuteness did not. A separate study of 191 people looking at pay, promotions and career satisfaction found that networking ability was the dimension that dominated. A one-year study of 135 career employees found that political skill predicted people's hierarchical position, income and career satisfaction a year later, and that the effect ran through their reputation.
Social astuteness still earns its keep, in a different place. A three-study investigation found that astuteness helped people *spot* the moment to take initiative when the climate invited it, and that interpersonal influence then helped them *turn* that initiative into good performance ratings. One part sees the opening; another walks through it.
Political skill also protects the person who has it. A study measuring anxiety, physical complaints, heart rate and blood pressure found that greater political skill reduced the effects of role conflict on every one of them. Another found that using political behaviour at work brought emotional labour for people low in political skill, while for people high in it the emotional labour did not appear. For people who spend all day managing competing demands from several senior people, that is not a small thing.
Now the finding that should make everyone slightly careful. A study of 173 supervisor and employee pairs looked at five impression management tactics, including self-promotion and ingratiation. People who used a lot of any of them and were politically skilled got better ratings than people who used them without the skill. And among people who used the tactics very little, those *without* political skill created a better impression than those with it. Skill turns tactics into results, and the tactics without the skill tend to backfire.
Then the part where good studies point in different directions. A study of 175 pairs in a government organisation found that when both kinds of fairness were low, political skill was positively related to performance, and when both were high it was *negatively* related. A two-country study in the United States and Greece found that political skill predicted performance where people saw little politics and made no difference where they saw a great deal. And a 2024 study found the opposite pattern: political skill combined with the will to use it predicted better performance and more citizenship in *more* political contexts, and not in less political ones. My reading is that these studies agree on the important thing, which is that political skill is not a fixed asset that pays out everywhere. It pays where the formal system leaves room, and the kind of room seems to matter as much as how much of it there is.
There is also an honest limit on how well any of this is understood. A 2015 review concluded that the studies support the effects of political skill without confirming how and why it produces them, and a 2020 overview noted that a comprehensive training programme has yet to be developed. What the research can tell you is which parts to practise; it cannot yet hand you a course.
Try this. Of the four parts, networking ability and apparent sincerity are the two most within your control this week. For networking, name one person two steps away from your executive who knows things you do not, and find a real work reason to speak to them. For sincerity, notice one moment this week when you are tempted to tell someone what they want to hear, and tell them what you think instead, kindly and with a reason.
The research behind this section
Ferris and colleagues (2005) developed and validated the Political Skill Inventory across three investigations involving seven samples, found a consistent four-factor structure, found political skill positively related to self-monitoring, political savvy and emotional intelligence, negatively related to trait anxiety and not correlated with general mental ability, and found that it predicted performance ratings of managers in two samples. Ferris and colleagues (2007) defined the construct domain of political skill and embedded it in a cognition, affect and behaviour framework for its effects on self and others. Munyon and colleagues (2015) meta-analysed political skill and found it positively related to self-efficacy, job satisfaction, organisational commitment, work productivity, citizenship, career success and personal reputation, negatively related to physiological strain, and not significantly related to psychological strain or perceptions of politics; it predicted task performance after controlling for the big five and general mental ability, partly through reputation and self-efficacy, and networking ability, interpersonal influence and apparent sincerity, but not social astuteness, predicted task performance. Bing and colleagues (2011) meta-analysed political skill as a predictor of task and contextual performance. Todd and colleagues (2009) found in 191 employees that political skill was associated with four of five career outcomes and that networking ability dominated the relationships. Blickle and colleagues (2011) collected two waves of data from 135 career employees over a year and found that political skill predicted hierarchical position, income and career satisfaction, with reputation mediating the relationships with position and satisfaction. Wei and colleagues (2010) found in 343 Chinese employees, their supervisors and 662 peers that supervisor-subordinate guanxi mediated the relationship between political skill and career development. Wihler and colleagues (2017) found across three studies that the social astuteness dimension interacted with climate for initiative to predict personal initiative, and that the interpersonal influence dimension interacted with initiative to predict supervisor-rated performance. Perrewé and colleagues (2004) found that greater political skill reduced the negative effects of role conflict on psychological anxiety, somatic complaints and physiological strain, including heart rate and blood pressure. Treadway and colleagues (2005) found in 193 employees that need for achievement and intrinsic motivation predicted political behaviour, and that emotional labour followed political behaviour for those low in political skill but was neutralised for those high in it. Harris and colleagues (2007) found in 173 supervisor-subordinate dyads that politically skilled people who used high levels of intimidation, exemplification, ingratiation, self-promotion or supplication received better supervisor ratings than unskilled users, and that at low usage unskilled people created a more desirable image than skilled ones. Andrews, Kacmar and Harris (2009) found in 175 dyads that political skill was positively related to performance when procedural and distributive justice were both low and negatively related when both were high. Kapoutsis and colleagues (2011) found in studies in the United States and Greece that political skill predicted job performance where perceived politics was low, with the effect attenuated where it was high. Frieder, Ellen and Kapoutsis (2024) found that political skill and political will interacted to predict higher task performance and citizenship in more political contexts but not less political ones. Brouer, Harris and Kacmar (2011) found mixed support in 106 dyads for self and manager political skill jointly moderating the effects of perceived politics. Kimura (2015) reviewed political skill research and concluded that empirical studies supported its theorised effects but failed to confirm how and why they occur. Kranefeld, Blickle and Meurs (2020) described the four facets and noted that a comprehensive training programme had yet to be developed.
Reading the informal organisation
Every organisation has two charts. One is on the intranet. The other is the pattern of who actually asks whom for advice, who trusts whom, and whose opinion gets sought before a decision is made, and it is the one the politics runs on.
In 1990 a researcher studied a small entrepreneurial firm and compared what people believed about the advice and friendship networks around them with what the networks actually were. People whose picture of the advice network was more accurate were rated as more powerful by their colleagues. Accuracy about the friendship network made no difference to how powerful people were seen to be. It is one of the most quoted findings in the network literature, and its message is precise: knowing who goes to whom for help with work is itself a form of power, over and above your formal position.
Formal and informal position matter as well. A 1984 study found that people's positions in the workflow, communication and friendship networks were strongly related to how influential colleagues thought they were and to who got promoted to supervisor. A later study found that network position and influence behaviour each related independently to how powerful people were seen to be. And a study in an advertising firm found that where people sat in the network shaped how they saw their own working conditions, which means your view of the organisation is partly a product of where you sit in it.
So who reads the map well? Studies in Italian research centres and a small firm found that accuracy depended on a person's position in the formal and informal structure and on personality, with people who were more central and closer to others seeing more. A set of experiments found that people learned a new network faster when they already knew that networks tend to have gaps in them, which suggests that the map can be learned like any other pattern.
Here, again, the evidence divides. A 2011 pair of experiments found that people primed to feel *low* in power perceived who was connected to whom more accurately than people primed to feel powerful, and that the difference disappeared for relationships that were not social. A 2018 field study in a call centre found that power was generally linked to *greater* accuracy, particularly about the relationships that mattered for getting the work done, and that accuracy was related to later transfers, promotions and departures. The experiments suggest that power makes people lazier observers; the field study suggests that in real jobs, people pay attention to the relationships they depend on. For an assistant, who depends on almost everyone and holds a good deal of informal power, both findings are a nudge to keep looking.
There are two systematic ways the map goes wrong. The first is status under pressure. A study combining a large national survey with an experiment found that when people felt their job was under threat, those with lower status called to mind smaller, tighter parts of their network, while those with higher status called to mind wider ones. Threat narrows the map for exactly the people who most need it wide. The second is gender. Across two studies, men were *perceived* to occupy the brokering roles in friendship networks more than women, even after taking account of their actual positions. Women who were misperceived as brokers were seen as competent but not warm, and where gender stereotypes were widely held, women performed worse on their individual tasks.
And there is the grapevine, which is part of the map whether anyone likes it or not. A network study found that friendship ties carried both positive and negative gossip, while workflow ties carried only positive gossip. The more someone gossiped, of either kind, the lower their supervisor rated their performance, and the higher their peers rated their informal influence. Gossip buys standing with colleagues at the price of standing with the boss, which is a trade an assistant cannot afford to make.
Knowing who asks whom for advice turned out to be a form of power in its own right, and it is the map an assistant is best placed to draw.
People who read these patterns well get a particular kind of reward. A series of studies found that high self-monitors, people who pay close attention to social cues and adjust to them, were more accurate about the status dynamics in their group, and raised their standing by becoming known as generous: they were sought out for help more and asked others for help less.
Try this. On one page, draw the advice network around your executive's next big decision: who they will actually ask before they decide, and who those people ask. Mark the arrows you are sure of in ink and the ones you are guessing in pencil. Over the next fortnight, check three of the pencil arrows by watching who is copied, who is called and who gets the early draft. The map will be wrong in places. That is fine; a map you correct is worth far more than one you carry in your head.
The research behind this section
Krackhardt (1990) studied a small entrepreneurial firm, compared perceptions of the friendship and advice networks with the actual networks, and found that those with more accurate cognitions of the advice network were rated as more powerful by others, while accuracy about the friendship network was not related to reputational power. Brass (1984) found that measures of employees' positions within workflow, communication and friendship networks were strongly related to supervisors' and non-supervisors' perceptions of influence and to promotions to supervisor, and that contacts beyond normal work requirements were particularly important for technical core personnel to acquire influence. Brass and Burkhardt (1993) found that structural position and behavioural tactics related independently and significantly to others' perceptions of an individual's power, each partially mediating the other. Ibarra and Andrews (1993) found in an advertising firm that advice network centrality and friendship network proximity shaped employees' perceptions of work-related conditions over and above individual attributes and formal positions. Casciaro (1998) studied three research centres in an Italian university and found that an observer's position in the formal and informal structure and personality traits both contributed to accuracy in network perception. Bondonio (1998) found in a small entrepreneurial firm that accuracy was predicted by degree centrality in the friendship or advice network and by demographic and social distance between workers. Janicik and Larrick (2005) found across five studies that prior knowledge of missing relations helped people learn an unfamiliar incomplete network and improved a strategic organisational choice. Simpson, Markovsky and Steketee (2011) found in experiments that participants primed with low power perceived who was tied to whom more accurately than those primed with high power, with no such difference for non-social relations. Marineau and colleagues (2018) found in a technical call centre that power was generally linked to increased perceptual accuracy, particularly for relationships most relevant to task and goal completion, and that network accuracy was related to subsequent transfer, promotion and exit. Smith, Menon and Thompson (2012) found using General Social Survey data and a laboratory experiment that, under job threat, people with low status activated smaller and tighter subsections of their networks while people with high status activated larger, less constrained ones. Brands and Kilduff (2014) found in an organisation and in MBA teams that men were perceived to occupy agentic brokerage roles in friendship networks more than women after controlling for actual positions, that women misperceived as brokers were seen as competent but not warm, and that widely endorsed gender stereotypes went with worse individual performance by women. Grosser, Lopez-Kidwell and Labianca (2010) found that friendship ties were related to both positive and negative gossip and workflow ties only to positive gossip, and that total gossip activity was negatively related to supervisor-rated performance and positively related to peers' evaluations of informal influence. Kurland and Pelled (2000) proposed a model of gossip and power in the workplace. Flynn and colleagues (2006) found across a series of studies that high self-monitors were more accurate in perceiving status dynamics and elevated their status by establishing a reputation as generous exchange partners, being sought out for help more and asking others for help less.
The hub, and the two ways to use it
Now to the position most Executive Assistants and Chiefs of Staff occupy without anyone ever calling it by its name. You sit between groups who do not talk to each other much: the executive and the teams, one director's world and another's, the board and the building. Network researchers call the gap between two groups a structural hole, and the person who spans it a broker.
The case for brokerage is strong. In a large American electronics company, pay, good performance evaluations, promotions and good ideas were disproportionately in the hands of people whose networks spanned the gaps between groups. Brokers were more likely to express ideas, less likely to have them dismissed and more likely to have them judged valuable, because opinion and behaviour are more alike within groups than between them, and people connected across groups can see options others cannot. An earlier study found that the value of this kind of network was highest for managers who had few peers doing the same work, because they had no crowd to take their bearings from. It is hard to think of a better description of an assistant to a chief executive.
The more recent research adds six refinements. The first is about how you broker. The classic image is the third party who benefits from keeping two others apart and playing them off against each other; the research calls this *tertius gaudens*, the third who enjoys. A 2005 study in an automotive engineering division proposed the opposite, *tertius iungens*, the third who joins: introducing people who do not know each other and helping people who do know each other work together. A joining orientation, dense networks and wide social knowledge each predicted involvement in innovation. A study of how brokers actually behave found that they did both: in short-lived exchanges outside their usual relationships they tended to hold the information themselves, and within their long-term relationships they tended to put people in direct touch. And a study in a global consumer products company found that an arbitraging orientation and a collaborating orientation changed how much value people extracted from the same brokerage position.
The second is about what it does to the people around you. A study of the French television game show industry over seventeen years looked at network hubs as a public good. Hubs brought novel ideas to the people near them, which helped those neighbours in creative roles; they also had so much cross-group activity that they could not commit fully to their neighbours' projects. A broker's gift and a broker's absence are the same position seen from two sides.
The third is about culture. A study of four high-tech companies in China found that bridging between disconnected contacts did not bring the career benefits seen in Western studies, and that the more an organisation had a close, high-commitment culture built on mutual investment, the more *harmful* structural holes were for people's salary and bonus. In a place built on sharing, holding the gap looks like hoarding.
The fourth is about trust, and it is the one I would pin to the wall. A 2018 study looked at people who sit between separate friendship cliques at work, and at what made their friends trust them. The obligations of friendship, openness and favouritism, pull against the organisation's need for discretion and neutrality. Brokers between cliques were trusted when they combined flexibility in how they presented themselves with a low tendency to blurt things out. For people inside a single close group, the opposite style, forthright and talkative, was what kept trust. Discretion is not a personality quirk in an assistant; it is what the position demands.
The fifth is a counterweight. A meta-analysis of 138 samples found that being sought out by many people, what researchers call indegree centrality, was more strongly related to job performance and career success than brokerage was. Being the person people come to may matter more than being the person between them.
The sixth is the cost. A study of communication networks found that people many colleagues depend on for information experienced more role overload and more role ambiguity, and through those, less of the vitality and learning researchers call thriving. Political skill reduced the ambiguity and did nothing for the overload. A theory paper made the same point about star performers, whose abundant connections were argued to produce information overload. Being the hub costs you, and skill will help you make sense of the demands without reducing their number.
A sociological review makes the ethical point directly: brokerage can ease interaction and help things happen, and it can also breed exploitation, the pursuit of personal gain and the accumulation of power.
Try this. Look back at the last week and list three times you passed information between two people or groups. For each, ask one question: did I leave them more connected to each other, or more dependent on me? Then choose one case where you could have introduced the two people directly, and do it next time, with a one-line note saying why they should talk.
The research behind this section
Burt (2004) studied managers in a large American electronics company and found that compensation, positive performance evaluations, promotions and good ideas were disproportionately held by people whose networks spanned structural holes, and that between-group brokers were more likely to express ideas, less likely to have them dismissed and more likely to have them evaluated as valuable. Burt (1997) found in a probability sample of senior managers that the value of social capital varied with the number of people doing the same work, being especially valuable to managers with few peers. Burt, Kilduff and Tasselli (2013) reviewed network advantage as a function of information breadth, timing and arbitrage. Obstfeld (2005) found in an engineering division of an automotive manufacturer that a tertius iungens orientation, dense social networks and diverse social knowledge predicted involvement in innovation, contrasting it with the tertius gaudens orientation of structural holes theory. Quintane and Carnabuci (2016) found using a relational event model that brokers more often brokered information through short-term interactions outside their long-term relationships, where they tended to intermediate consistent with a tertius gaudens strategy, while through long-term ties they tended to facilitate direct exchange consistent with a tertius iungens strategy. Soda, Tortoriello and Iorio (2018) found in a global consumer product company that arbitraging and collaborating orientations had differential effects on the relationship between brokerage and performance. Clement, Shipilov and Galunic (2018) studied the French television game show industry from 1995 to 2012 and found that network hubs created positive externalities by bringing novel ideas to network neighbours in creativity-focused roles, and negative externalities because heavy cross-community demands limited their commitment to neighbours' projects. Xiao and Tsui (2007) studied four high-tech companies in China and found that structural holes did not confer the career benefits found in Western contexts and were more detrimental to salary and bonus the more the organisation had a clan-like, high-commitment culture. Tasselli and Kilduff (2018) found in two samples that Simmelian brokers between friendship cliques were trusted by their friends when they combined high self-monitoring with low blirtatiousness, while non-brokers embedded in a single clique were trusted with the opposite, forthright and loquacious style. Fang and colleagues (2015) meta-analysed 138 samples and found that self-monitoring predicted indegree centrality and brokerage, that personality explained 3% to 5% of the variance in network position, and that indegree centrality was more strongly related to job performance and career success than brokerage. Cullen, Gerbasi and Chrobot-Mason (2018) found an indirect cost of communication network centrality on thriving through role overload and role ambiguity, and found that political skill mitigated role ambiguity but not role overload. Oldroyd and Morris (2012) theorised that star employees' abundant social capital may produce information overload. Gould and Fernandez (1989) set out five formally distinct brokerage roles in transaction networks. Halevy, Halali and Zlatev (2019) reviewed brokerage and brokering and described brokering as a social influence process of intermediation or modification that can affect others' relationships positively or negatively. Kwon and colleagues (2020) reviewed the network brokerage literature across disciplines. Stovel and Shaw (2012) described the dual aspect of brokerage, which can ease social interaction and facilitate economic and political activity but often breeds exploitation, personal profit, corruption and the accumulation of power.
Self-serving or pro-social: where the line is
The original worry about politics is that it is about looking after yourself at other people's expense. The research has a name for the motive behind political behaviour, political will, and it has started to separate the two versions.
A four-study project across the United States, Greece and the United Kingdom, with 925 people, built a scale for political will with a self-serving side and a benevolent side, and found that political will predicted influence behaviour, status and career growth potential over and above political skill. Wanting to do it adds something to being able to do it.
The honest part came when a later study tested that scale with ratings from several people who knew each person. The self-serving side held up: it related to striving for power, and to people's hierarchical position and income. The benevolent side did not hold up as a measure of truly altruistic motives, and the authors called for new items that capture benevolence towards others at work. In other words, research has found it easier to measure selfish political motivation than the generous kind. That is a gap in the evidence, and it should not be mistaken for evidence that the generous kind does not exist.
What makes people willing to engage at all? Interviews with fourteen managers in a global semiconductor company found three kinds of ambivalence behind political will: functional, about whether politics works; ethical, about whether it is right; and emotional, about how it feels. A later study of leaders at different levels described a journey from politically naive to politically mature in three stages, involving changes in mindset as well as skill, and found that experience and significant others, mentors and colleagues who showed the way, were what moved people along.
The most useful finding on motives comes from the research on helping at work. Two field studies found that people who were both "good soldiers", motivated to help, and "good actors", motivated to look good, were the most likely to help colleagues, be courteous and take initiative. Impression management strengthened the link between wanting to help and helping in ways that were visible. There was one exception, and it matters. Only the genuine wish to help predicted voice, the harder form of citizenship that means speaking up with a concern or a better idea. Looking good and doing good can travel together; speaking up takes the real thing.
At the far end of self-serving politics sits Machiavellianism, the disposition to manipulate. A review of the psychological and evolutionary literature concluded that it does not correlate with general intelligence and does not consistently lead to real-world success. A meta-analysis of 245 samples and 43,907 people found that Machiavellianism and psychopathy were consistently associated with lower job performance and that all three dark traits were associated with counterproductive behaviour.
And now the uncomfortable finding about charm. A 2020 study with 1,438 people found that political skill effectively masked Machiavellianism: highly Machiavellian people who were politically skilled received high performance ratings from colleagues. Only among those with low political skill did colleagues eventually rate the Machiavellians poorly, and then only after long tenure. The same study found that the skilled Machiavellians did not hold back from mean behaviour or from taking organisational resources early on. The "apparent" in apparent sincerity is doing real work. It is why I think the only reliable test of pro-social politics is what it does over time: who ends up better off, and whether they would be happy to hear how it happened.
The test of good politics is whether everyone involved would be comfortable hearing exactly how it was done.
There is a brighter side to the same skills. Leader political skill has been found to shape employees' sense of being supported and their trust, and politically skilled leaders understood situations better and showed their people more consideration. Researchers have even proposed a construct called leader political support, for the ways leaders use political skill on behalf of the people who work for them. And the skilled appear to be selective: political skill and will predicted which combination of impression management tactics people chose.
Try this. Before your next political move, however small, whether it is a word in someone's ear or deciding who gets the draft first, run the three-question check. Who benefits, and would the organisation recognise the benefit as its own? Is anyone worse off who does not know it yet? Would I be comfortable if the other person heard exactly how this was done? If the answers are good, go ahead with a clear conscience. If they are not, change the move.
The research behind this section
Kapoutsis and colleagues (2017) conducted four studies with six samples totalling 925 people in the United States, Greece and the United Kingdom, established the Political Will Scale, and found that political will related positively to influence and work behaviours and explained variance over and above political skill in influence tactics, status and career growth potential. Blickle, Schütte and Wihler (2018) used a triadic multisource design and found support for the construct validity (power striving) and criterion validity (hierarchical position and income) of the self-serving political will scale, but did not support the interpretation of the benevolent scale as representing an altruistic political motive. Treadway and colleagues (2005) used Mintzberg's conceptualisation of political will and skill and found that need for achievement and intrinsic motivation predicted political behaviour. Doldor, Anderson and Vinnicombe (2013) interviewed 14 managers in a global semiconductor company and identified three attitudinal ambivalences in managerial political will, functional, ethical and emotional, and examined gender differences in a masculine organisational setting. Doldor (2017) interviewed leaders at different seniority levels and proposed a three-stage model of political maturation involving changes in mindsets and cognitive scripts as well as skills, facilitated by experience and significant others. Grant and Mayer (2009) found in two field samples that impression management motives strengthened the association between prosocial motives and affiliative citizenship towards people and the organisation, and that only prosocial motives predicted voice. Wilson, Near and Miller (1996) concluded that Machiavellianism does not correlate with general intelligence and does not consistently lead to real-world success. O'Boyle and colleagues (2012) meta-analysed 245 samples (N = 43,907) and found that reductions in job performance were consistently associated with Machiavellianism and psychopathy, and that counterproductive work behaviour was associated with all three dark triad traits, moderated by authority and culture. Blickle and colleagues (2020) studied 1,438 participants in two multisource studies and found that when political skill was high, targets received high career role performance ratings from coworkers even when high in Machiavellianism, that high Machiavellians with low political skill were rated low at long tenure, and that people high in both did not avoid overtly mean behaviours or extracting organisational resources at short tenure. Treadway and colleagues (2004) found support for effects of leader political skill on employees' perceived support, trust and other reactions. Frieder and colleagues (2019) found that leader understanding mediated the relationship between political skill and consideration behaviours, and that consideration was positively related to subordinates' perceptions of leader instrumentality. Ellen, Ferris and Buckley (2013) proposed leader political support as a reconsideration of leader political behaviour, and Ellen (2014) set out its positive possibilities. Maher and colleagues (2018) found across multiple studies that political skill and political will predicted the selection of impression management configurations. Brass, Butterfield and Skaggs (1998) set out a social network perspective on relationships and unethical behaviour.
Moving an idea through a political place
Most of the political work an assistant does is in service of something: an idea of your own, a change your executive wants, a decision that needs to land well. So what does the research say about how things actually get through?
Start with a reassurance about decisions. A study of 61 strategic decisions in 24 companies found that rational process, gathering information and analysing it, and political behaviour were independent dimensions. A decision could be highly rational and highly political at once, or neither. Politics is not the opposite of good process; it runs alongside it.
Whether that politics helps or harms decisions is contested, and I would rather show you both sides than pretend otherwise. Most of the research treats political behaviour in decisions as damaging. A study of 228 middle managers in 114 Greek firms found that when politics was high during implementation, implementation success fell, and that trust in the top team reduced the damage. On the other side, a study of 200 respondents in Dubai found that constructive politics contributed to decision success and speed, and a study of 103 product development teams found that political behaviour was positively related to speed to market. A study of 117 strategic decisions found a middle path: top teams that were behaviourally integrated and shared power more widely were better able to limit the damaging effects of politics on the quality of their decision process. The pattern that emerges is that politics goes wrong in fragmented, low-trust teams and can speed things up in cohesive ones.
On change, the best evidence comes from 68 change initiatives in the United Kingdom's National Health Service. A change agent whose network bridged disconnected groups was better placed to start and push through changes that departed sharply from how things were done, and *worse* placed to get smaller, less divergent changes adopted, which were helped more by a close-knit network. A second study of the same initiatives looked at resistance. Strong ties to influential people who were ambivalent about a change, the fence-sitters, made adoption more likely. Strong ties to outright opponents helped only when the change was modest; with bigger changes they became a liability.
That gives you a practical rule. For a small change, work through the people you are close to, including the sceptics. For a big one, reach across groups for ideas and allies, spend your warmth on the undecided, and do not assume a friendship with an opponent will carry a radical proposal.
Interviews with change agents add the texture: political behaviour was an accepted part of the role, and it served organisational goals such as protecting a change agenda as well as personal careers. And the skill is not only individual. A study of 189 student teams and 28 work teams found that a team's collective political skill predicted its performance through task and social cohesion, with no sign that too much of it did harm.
Try this. For the next idea you want to move, write three lists: the people who will decide, the people they will ask, and the people who are undecided. Then decide how divergent your idea is. If it is modest, talk it through with the sceptic you get on with first. If it is a real departure, spend your time with the undecided and with people outside your usual circle, and bring their words into the paper.
The research behind this section
Dean and Sharfman (1993) studied 61 strategic decisions in 24 companies using multiple informants and found that procedural rationality and political behaviour were independent dimensions of strategic decision-making. Lampaki and Papadakis (2018) found in 228 middle managers from 114 private Greek firms that high organisational politics during implementation reduced implementation success, and that trust in the top management team reduced the negative impact. Elbanna (2018) found in 200 respondents in Dubai that constructive politics contributed to decision success and decision pace. Dayan, Elbanna and Di Benedetto (2012) found in 103 Turkish new product development teams that political behaviour positively influenced speed to market. Shepherd and colleagues (2020) studied 117 strategic decisions and found that behaviourally integrated and decentralised top management teams were better equipped to attenuate the damaging effects of political behaviour on the quality of their decision processes. Battilana and Casciaro (2012) analysed 68 change initiatives in the National Health Service and found that low structural closure in a change agent's network aided the initiation and adoption of changes that diverged from the institutional status quo but hindered the adoption of less divergent changes. Battilana and Casciaro (2013) found in the same initiatives that strong ties to influential fence-sitters increased the probability of adoption, while strong ties to resistors helped only when a change diverged little from institutionalised practice and could become liabilities with more divergent changes. Buchanan and Badham (1999) drew on interviews with change agents and found that political behaviour was an accepted dimension of the change agency role, that change agents were drawn into it by organisational and interpersonal factors, and that it served organisational goals as well as personal career objectives. Lvina, Johns and Vandenberghe (2018) found in 189 student project teams and 28 business work teams that team political skill related to team performance through task and social cohesion, and did not support a too-much-of-a-good-thing effect.
The political read
Everything above comes down to a moment most assistants know well: there is something you want to move, a decision, a change or an idea, and you need to work out where the resistance is and what to do about it without compromising yourself.
Put in what you are trying to move, how political the place feels, how well you know the advice network around the decision, how big a departure it is, where you sit, and whose interests it serves. It will give you a reading, three things to look at, the steps the research points to, what to check afterwards, and a one-page map to copy.
Where to start
If you take one thing from all of this, make it the map. The research is remarkably consistent that knowing who asks whom for advice is a source of influence in its own right, that the people best placed to draw that map are the ones at the centre of the information flow, and that the map can be learned, checked and corrected like any other piece of knowledge.
After that, there are three moves worth making.
Share the understanding: when a decision is coming, tell people who decides, when and on what, because information is what the research keeps finding makes a place feel less political. Join people up: when you pass something between two groups, leave them more connected to each other than to you. And run the three-question check before any move: who benefits, who is worse off without knowing it, and whether you would be comfortable with them hearing exactly how it was done.
I have put the three pages I would use into a download: an informal map for any decision that matters, a coalition sheet for moving an idea through the people who decide and the people who are undecided, and a broker's ledger for keeping an honest record of what you pass on and why.
Watching from the sidelines taught me that politics was happening whether I took part or not. The research has taught me something kinder: that the people who do it best are usually the ones who understand the organisation most accurately, connect people generously and can be trusted with what they know. That describes a great many assistants already, and it is good to see the evidence say so.
The part of this that shows up in how you are seen at work sits inside what the Connected Leader Assessment calls Influence and Stakeholder Trust: how well you read the people around your executive, how you build support for decisions and ideas, and whether people trust you with what passes through your hands. It is the dimension that decides whether your position at the hub becomes influence or only work.
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 look at which of your relationships and which parts of your informal map the organisation could not replace tomorrow, which is exactly the influence most worth using well.
Stay savvy.
Meg ✌️
Free tools
The political read, the interactive tool in this post, free
The Politics Pages, the download in this post, free
The Connected Leader Assessment, twelve minutes, free, with a read on Influence and Stakeholder Trust, the dimension this post sits in
P.S. Want to Go Deeper on Office Politics?
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 influence, informal networks and using a seat at the centre of things well, 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.
📃 Buchanan (2008) 250 British managers. Politics was common, mostly seen as ethical and necessary, and 80% had no training in it. https://doi.org/10.1111/j.1467-8551.2007.00533.x
📃 Ferris and colleagues (2002) Where research on perceptions of organisational politics stood, and where it should go. https://doi.org/10.1016/s1475-9144(02)01034-2
📃 Chang, Rosen and Levy (2009) Meta-analysis. Perceived politics went with strain, intentions to quit and lower satisfaction, and acted as a hindrance stressor. https://doi.org/10.5465/amj.2009.43670894
📃 Miller, Rutherford and Kolodinsky (2008) 79 samples, 25,059 people. Strong links with satisfaction and commitment; none with in-role performance. https://doi.org/10.1007/s10869-008-9061-5
📃 Bedi and Schat (2013) 118 samples, 44,560 people. Politics went with lower trust, more stress and burnout, and lower performance. https://doi.org/10.1037/a0034549
📃 Kacmar and Carlson (1997) The perceptions of politics scale tested and refined across 2,758 respondents. https://doi.org/10.1177/014920639702300502
📃 Ferris and Kacmar (1992) Feedback, autonomy, skill variety and promotion chances all shaped how political a workplace felt. https://doi.org/10.1177/014920639201800107
📃 Kacmar and colleagues (1999) 1,255 employees. Understanding softened the link between politics and satisfaction, and nothing else. https://doi.org/10.1177/001872679905200305
📃 Hochwarter and colleagues (2020) Thirty years of research on how political a workplace feels, reviewed. https://doi.org/10.1177/0149206319898506
📃 Rosen, Levy and Hall (2006) 150 pairs. A good informal feedback culture went with seeing less politics, and better morale and performance. https://doi.org/10.1037/0021-9010.91.1.211
📃 Vigoda (2002) Three samples. Distress was an immediate response to politics and helped explain later aggression. https://doi.org/10.1002/job.160
📃 Andrews and Kacmar (2001) 418 employees. Politics, fairness and organisational support were distinct from one another. https://doi.org/10.1002/job.92
📃 Landells and Albrecht (2017) Interviews with 14 employees. Four ways of seeing politics: reactive, reluctant, strategic and integrated. https://doi.org/10.1007/s10869-015-9434-5
📃 Maslyn, Farmer and Bettenhausen (2017) Moderate negative politics was judged better than a lot or none; positive politics, the more the better. https://doi.org/10.1177/0018726717704706
📃 Mintzberg (1985) The organisation as a political arena, and the functional roles politics plays. https://doi.org/10.1111/j.1467-6486.1985.tb00069.x
📃 Buchanan and Badham (1999) Change agents described politics as an accepted part of the role, serving the organisation as well as careers. https://doi.org/10.1177/001872679905200503
📃 Fedor and colleagues (2008) 119 people. Positive and negative politics were separate perceptions, and the positive kind added to satisfaction. https://doi.org/10.1111/j.1559-1816.2008.00297.x
📃 Ellen (2014) The positive possibilities of leaders' political behaviour. https://doi.org/10.1002/job.1938
📃 Ellen, Ferris and Buckley (2013) Leader political support: leaders using political skill on behalf of their people. https://doi.org/10.1016/j.leaqua.2013.10.007
📃 Doldor, Anderson and Vinnicombe (2013) Interviews with 14 managers. Political will held three ambivalences: functional, ethical and emotional. https://doi.org/10.1111/j.1467-8551.2012.00812.x
📃 Doldor (2017) Leaders moved from politically naive to politically mature in three stages, helped by experience and others. https://doi.org/10.1111/1467-8551.12219
📃 Kapoutsis and colleagues (2017) 925 people, three countries. Political will predicted influence, status and career growth beyond political skill. https://doi.org/10.1177/0149206314566460
📃 Treadway and colleagues (2005) 193 employees. Political behaviour cost the unskilled emotional labour; the skilled did not pay it. https://doi.org/10.1002/job.310
📃 Ferris and colleagues (2007) The construct of political skill defined, with a framework for how it works. https://doi.org/10.1177/0149206307300813
📃 Bing and colleagues (2011) Meta-analysis of political skill as a predictor of task and contextual performance. https://doi.org/10.1016/j.jvb.2011.02.006
📃 Perrewé and colleagues (2004) Political skill reduced the effect of role conflict on anxiety, heart rate and blood pressure. https://doi.org/10.2307/20159566
📃 Treadway and colleagues (2004) Leaders' political skill shaped employees' sense of support and their trust. https://doi.org/10.1016/j.leaqua.2004.05.004
📃 Todd and colleagues (2009) 191 employees. Political skill predicted four of five career outcomes; networking ability led. https://doi.org/10.3200/socp.149.3.279-304
📃 Harris and colleagues (2007) 173 pairs. Impression management worked for the politically skilled and backfired for the unskilled. https://doi.org/10.1037/0021-9010.92.1.278
📃 Ferris and colleagues (2019) A review proposing a broader way to organise research on organisational politics. https://doi.org/10.1146/annurev-orgpsych-012218-015221
📃 Ellen and colleagues (2022) 177 employees. An inverted U between politics and satisfaction, and for loyal staff, performance. https://doi.org/10.1111/apps.12347
📃 Andrews, Kacmar and Harris (2009) 175 pairs. Political skill helped performance when fairness was low and hurt it when fairness was high. https://doi.org/10.1037/a0017154
📃 Brouer, Harris and Kacmar (2011) 106 pairs. Mixed support for political skill softening the effects of perceived politics. https://doi.org/10.1002/job.718
📃 Kimura (2015) Review. The effects of political skill are supported; how and why they happen is not yet shown. https://doi.org/10.1111/ijmr.12041
📃 Cullen, Gerbasi and Chrobot-Mason (2018) Being central in communication brought overload and ambiguity; political skill eased only the ambiguity. https://doi.org/10.1177/0149206315571154
📃 Frieder and colleagues (2019) Politically skilled leaders read situations better and showed more consideration. https://doi.org/10.1111/peps.12336
📃 Blickle, Schütte and Wihler (2018) Self-serving political will predicted position and income; the benevolent scale did not hold up as altruism. https://doi.org/10.1016/j.jvb.2018.03.002
📃 Wihler and colleagues (2017) Three studies. Social astuteness spotted the opening for initiative; interpersonal influence turned it into ratings. https://doi.org/10.1177/0149206314552451
📃 Blickle and colleagues (2011) 135 employees over a year. Political skill predicted position, income and satisfaction, through reputation. https://doi.org/10.1111/j.1559-1816.2011.00862.x
📃 Maher and colleagues (2018) Political skill and will predicted which combination of impression management tactics people chose. https://doi.org/10.1016/j.jvb.2018.05.004
📃 O'Boyle and colleagues (2012) 245 samples, 43,907 people. Machiavellianism went with lower performance and more counterproductive behaviour. https://doi.org/10.1037/a0025679
📃 Grant and Mayer (2009) Good soldiers who were also good actors helped most; only the wish to help predicted speaking up. https://doi.org/10.1037/a0013770
📃 Wilson, Near and Miller (1996) Machiavellianism was unrelated to intelligence and did not consistently bring success. https://doi.org/10.1037/0033-2909.119.2.285
📃 Blickle and colleagues (2020) 1,438 people. Political skill masked Machiavellianism in colleagues' ratings. https://doi.org/10.1016/j.jvb.2020.103401
📃 Wei and colleagues (2010) 343 employees in China. Political skill reached career prospects through the relationship with the boss. https://doi.org/10.1111/j.1467-6486.2009.00871.x
📃 Munyon and colleagues (2015) Meta-analysis. Political skill predicted performance beyond personality and intelligence; astuteness alone did not. https://doi.org/10.1111/peps.12066
📃 Ferris and colleagues (2005) The Political Skill Inventory: four parts, unrelated to intelligence, predictive of managers' ratings. https://doi.org/10.1177/0149206304271386
📃 Frieder, Ellen and Kapoutsis (2024) Political skill with the will to use it predicted performance in more political settings, not less. https://doi.org/10.1037/apl0001107
📃 Kapoutsis and colleagues (2011) United States and Greece. Political skill predicted performance where politics was low, not where it was high. https://doi.org/10.1016/j.jvb.2010.09.009
📃 Vigoda (2000) 303 public sector employees. Politics brought neglect more than leaving, and a weak dip in performance. https://doi.org/10.1006/jvbe.1999.1742
📃 Abbas and colleagues (2014) 231 employees. Psychological capital softened politics' effects, and made people readier to leave. https://doi.org/10.1177/0149206312455243
📃 Yang (2017) 577 people, 133 units. Politics hurt performance in loose structures and helped it in tight ones. https://doi.org/10.1080/1359432x.2016.1251417
📃 Krackhardt (1990) A small firm. People who read the advice network accurately were seen as more powerful. https://doi.org/10.2307/2393394
📃 Brass (1984) Network position predicted influence and promotion, and extra contacts mattered most in technical roles. https://doi.org/10.2307/2392937
📃 Burt (2004) Brokers across groups earned more, were promoted faster and had more ideas judged valuable. https://doi.org/10.1086/421787
📃 Obstfeld (2005) The third who joins: connecting people predicted involvement in innovation. https://doi.org/10.2189/asqu.2005.50.1.100
📃 Casciaro (1998) Position and personality both shaped how accurately people saw the network around them. https://doi.org/10.1016/s0378-8733(98)00008-2
📃 Brands and Kilduff (2014) Men were seen as brokers more than women in the same positions, with costs for the women. https://doi.org/10.1287/orsc.2013.0880
📃 Tasselli and Kilduff (2018) Brokers between friendship groups were trusted when adaptable and not given to blurting things out. https://doi.org/10.5465/amj.2015.0856
📃 Flynn and colleagues (2006) High self-monitors read status dynamics better and rose by being known as generous helpers. https://doi.org/10.1037/0022-3514.91.6.1123
📃 Simpson, Markovsky and Steketee (2011) Experiments. People primed to feel less powerful saw who was tied to whom more accurately. https://doi.org/10.1016/j.socnet.2010.10.007
📃 Marineau and colleagues (2018) A call centre. Power went with greater accuracy about task relationships, which predicted later careers. https://doi.org/10.1016/j.socnet.2018.01.006
📃 Smith, Menon and Thompson (2012) Under job threat, low-status people called a narrower network to mind; high-status people a wider one. https://doi.org/10.1287/orsc.1100.0643
📃 Brass and Burkhardt (1993) Network position and influence behaviour each shaped how powerful people were seen to be. https://doi.org/10.2307/256588
📃 Ibarra and Andrews (1993) Where people sat in the network shaped how they saw their working conditions. https://doi.org/10.2307/2393414
📃 Battilana and Casciaro (2012) 68 NHS changes. Bridging networks helped radical changes and hindered modest ones. https://doi.org/10.5465/amj.2009.0891
📃 Battilana and Casciaro (2013) 68 NHS changes. Strong ties to fence-sitters helped; ties to resistors helped only small changes. https://doi.org/10.1287/mnsc.1120.1583
📃 Clement, Shipilov and Galunic (2018) Network hubs brought ideas to their neighbours, and less commitment to their projects. https://doi.org/10.1177/0001839217708984
📃 Xiao and Tsui (2007) Four Chinese firms. In high-commitment cultures, holding the gap between groups hurt pay and bonus. https://doi.org/10.2189/asqu.52.1.1
📃 Quintane and Carnabuci (2016) Brokers held information in passing exchanges and joined people up in long-term relationships. https://doi.org/10.1287/orsc.2016.1091
📃 Halevy, Halali and Zlatev (2019) Review of brokerage and brokering: how people change others' relationships, for better or worse. https://doi.org/10.5465/annals.2017.0024
📃 Kwon and colleagues (2020) An integrative review of network brokerage research. https://doi.org/10.1177/0149206320914694
📃 Fang and colleagues (2015) 138 samples. Being sought out mattered more for performance and careers than brokering. https://doi.org/10.1287/orsc.2015.0972
📃 Oldroyd and Morris (2012) Stars' abundant connections can produce information overload. https://doi.org/10.5465/amr.2010.0403
📃 Grosser, Lopez-Kidwell and Labianca (2010) Gossip went with lower ratings from the boss and more informal influence with peers. https://doi.org/10.1177/1059601109360391
📃 Kurland and Pelled (2000) A model of gossip and power in the workplace. https://doi.org/10.2307/259023
📃 Ferris and colleagues (1996) The perceptions of politics model supported, with control and understanding softening its effects. https://doi.org/10.1177/001872679604900206
📃 Burt, Kilduff and Tasselli (2013) Network advantage reviewed: information breadth, timing and arbitrage. https://doi.org/10.1146/annurev-psych-113011-143828
📃 Burt (1997) Bridging contacts was worth most to managers with few peers doing the same work. https://doi.org/10.2307/2393923
📃 Soda, Tortoriello and Iorio (2018) Arbitraging and collaborating brokers got different returns from the same position. https://doi.org/10.5465/amj.2016.0123
📃 Janicik and Larrick (2005) People who expected gaps in a network learned it faster. https://doi.org/10.1037/0022-3514.88.2.348
📃 Brass, Butterfield and Skaggs (1998) A social network perspective on relationships and unethical behaviour. https://doi.org/10.2307/259097
📃 Gould and Fernandez (1989) Five formally distinct brokerage roles in networks. https://doi.org/10.2307/270949
📃 Stovel and Shaw (2012) Brokerage eases interaction and can also breed exploitation and the accumulation of power. https://doi.org/10.1146/annurev-soc-081309-150054
📃 Bondonio (1998) Centrality and closeness to colleagues predicted accuracy in seeing informal networks. https://doi.org/10.1016/s0378-8733(98)00007-0
📃 Shepherd and colleagues (2020) 117 decisions. Integrated, decentralised top teams limited the damage politics did. https://doi.org/10.1016/j.lrp.2019.101943
📃 Lampaki and Papadakis (2018) 228 middle managers. Politics lowered implementation success; trust in the top team softened it. https://doi.org/10.1016/j.emj.2018.07.005
📃 Elbanna (2018) 200 respondents. Constructive politics contributed to decision success and speed. https://doi.org/10.1016/j.emj.2018.06.006
📃 Lvina, Johns and Vandenberghe (2018) Team political skill predicted team performance through cohesion. https://doi.org/10.1177/0149206315598371
📃 Dean and Sharfman (1993) 61 decisions. Rational process and political behaviour were independent of each other. https://doi.org/10.1111/j.1540-5915.1993.tb00504.x
📃 Dayan, Elbanna and Di Benedetto (2012) 103 product teams. Political behaviour went with faster speed to market. https://doi.org/10.1109/tem.2011.2166078
📃 Fu and Peng (2025) People who approach challenges built relationships in political settings; avoiders withdrew and did worse. https://doi.org/10.1111/apps.70027
📃 Kranefeld, Blickle and Meurs (2020) Political skill's four facets, and the note that a full training programme is yet to be built. https://doi.org/10.1093/acrefore/9780190236557.013.747
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