The Perils and Promises of Self-Disclosure on Social Media

In addition to their professional social media accounts, individuals are increasingly using their personal profiles and casual posts to communicate their identities to work colleagues. They do this in order to ‘stand out from the crowd’ and to signal attributes that are difficult to showcase explicitly in a work setting. Existing studies have tended to treat personal posts viewed in a professional context as a problem, since they can threaten impression management efforts. These accounts focus on the attempts of individuals to separate their life domains on social media. In contrast, we present the narratives of professional IT workers in India who intentionally disrupt the boundaries between personal and professional profiles in order to get noticed by their employers. Drawing on the dramaturgical vocabulary of Goffman (1959) we shed light on how individuals cope with increased levels of self-disclosure on social media. We argue that their self-presentations can be likened to post-modern performances in which the traditional boundaries between actor and audience are intentionally unsettled. These casual posts communicate additional personal traits that are not otherwise included in professional presentations. Since there are no strict boundaries between formal front-stage and relaxed back-stage regions in these types of performance, a liminal mental state is often used, which enables a better assessment of the type of information to present on social media.


Introduction
Social media enable modes of professional self-presentation that were previously impossible.By carefully constructing social media profiles and posts otherwise invisible information can be shared (Leonardi 2014).Such material can include posts about an individual's expertise, the extent of their social networks and their association with specific projects, which can help make them more noticeable and facilitate new opportunities (Bharati et al. 2014;Treem and Leonardi 2012).With these end goals in mind individuals give extensive thought to what they share in order to ensure that the inferences fostered via social media are favourable.The cumulative impression created by profiles and posts can be used in a variety of ways to make judgements that can have significant implications for their career trajectories (Roth et al. 2016).
The growing attention employers give to casual posts from personal profiles has caused some uneasiness (Clark and Roberts 2010).Increasingly, the content of informal posts is used alongside professional information to form inferences about a person and their capabilities (Neeley and Leonardi 2016).Judgements can be informed by personal profiles, posts about out of work activities and information or pictures posted by other people (Neeley and Leonardi 2016).The ubiquity of the internet and the interconnectedness of social media networks mean that professional and social boundaries are blurred, resulting in a spill over of information from one life domain to another (French and Read 2013;Richey et al. 2016).As information seeps across digital networks it becomes more difficult to control impressions cultivated by social media.The negative inferences associated with contextually inappropriate personal posts can have serious consequences for individuals and organisations (Richey et al. 2016).For instance, Justine Sacco became infamous for a distasteful joke on her personal Twitter account, posted as 2 Professional Self-Presentations on Social Media Social media are increasingly being used by individuals to create positive professional impressions (Leonardi 2014;Ollier-Malaterre et al. 2013).This process is initiated as profiles are made visible to work colleagues and professional connections, enabling them to traverse a variety of different personal information (Treem and Leonardi 2012).Social media footprints are not limited to a single account and often comprise information distributed across a number of different sites (Vaast 2007).Colleagues may simultaneously be connected on LinkedIn and Facebook, view each other's tweets and blog posts and be part of a bespoke organisational wiki or network.Each source of information helps make inferences about the attributes and characteristics of a person.In a professional setting such inferences can inform decisions about hiring and career advancement (Wilson et al. 2012;Bohnert and Ross 2010).For this reason social media accounts are carefully curated in order to highlight expertise (Treem and Leonardi 2012) and to enhance likeability (Forest and Wood 2012;Hollenbaugh and Ferris 2014).Allowing professional contacts to access social media posts is a form of self-disclosure, which can both promote and damage professional relationships (Chen and Sharma 2013;Cuddy et al. 2011;Dutta 2010).Many social media studies adopt the Bideology of openness^, which suggests that effective communication relies upon the maximum possible degree of transparency (Eisenberg and Witten 1987).In this view organisations can reap substantial benefits if actors are willing to share information about themselves and their work (Leonardi 2014).In fact, it is what social media conceals as well as what it reveals that can be of great advantage in a professional setting (Scott and Orlikowski 2014).The dominant assumption in social media studies is that the platforms provide actors with a degree of control over how they are perceived by their professional networks (Leonardi 2014).
A major advantage for professional users is that information presented can be carefully considered and edited, allowing individuals additional control over what is communicated (Walther 2007).Conversation threads and posts remain visible long after an interaction has taken place and can be viewed by many others, making careful deliberation an even more pressing issue (Treem and Leonardi 2012).The enduring nature of posts can be especially advantageous in a professional setting where they may be accessed over an extended period of time by many intended and unexpected viewers (Leonardi 2014).In large, geographically dispersed organisations these networks have fostered connections and collaborations that may not have occurred otherwise (Turner and Reinsch 2010;Mount and Martinez 2014).As well as giving individuals the opportunity to transcend the usual limits of time and space, the disparate sources of information highlight characteristics that might otherwise be challenging for individuals to share in a professional setting (Turner and Reinsch 2010;Ollier-Malaterre et al. 2013).Attributes that might be viewed favourably on their own do not always sit comfortably together.For example, a person may wish to be thought of as both competitive and warm (Ollier-Malaterre et al. 2013).Social media can communicate characteristics implicitly by inference, removing some of the burden placed on individuals to explicitly exhibit juxtaposing traits.
The challenges of communicating in this way emerge because it can be difficult to control the types of information made available via social media.Individuals are not the only ones contributing information to their social media accounts.They can also be tagged in photographs or posts added by other people (Utz 2010).Algorithms also selectively curate what can be viewed.These circumstances increase the possibility that information from one life domain may spill over into another (Richey et al. 2016;French and Read 2013).In concert these bits of information help to infer what a person is like (Utz 2010).Managers have been shown to traverse personal social media accounts to find out about potential employees and to check the behaviour of current hires (Van Iddekinge et al. 2016;Roth et al. 2016).Social media snooping becomes problematic if the information made visible to professional contacts alters favourable perceptions of a person.There are many examples of employees that have been fired after posts made to private accounts became visible to their employers.A classic example is that of the employee who posts disparaging remarks about their boss on social media, only to be fired when the boss finds those posts (LaFerla 2006).Different cultural and organisational contexts render different behaviours contextually inappropriate, making it very challenging for individuals to anticipate the standards by which their posts might be scrutinized (Richey et al. 2016).If the social cues that inform interaction and behaviour are not visible or if they are misjudged then inappropriate posts can be made visible to colleagues, altering their perceptions of an individual (Richey et al. 2016;Spottswood and Hancock 2017).Indeed, multiple contradictory selves can seem to exist in a Bfragmented electronically mediated environment^ (Vaast 2007).
Different coping strategies are used to minimise these potential threats.Individuals employ a range of boundary management techniques to try and prevent their personal lives from spilling into the professional domain (Ollier-Malaterre et al. 2013).These include self-censoring (Lampinen et al. 2009), modifying privacy settings (French and Read 2013) and the creation of multiple profiles (Stutzman and Hartzog 2012).Current studies focus squarely on efforts to separate life domains and manage social boundaries on social media, but thus far give little attention to the intentional use of personal social media content in a professional setting.We argue that individuals are increasingly creative about their online self-presentations because they want to stand out.Our findings show how information from personal and professional social media accounts is drawn together to signal favourable personal characteristics to others at work.Following other social media studies, we turn to the sociological literature for a theoretical explanation of how social media posts are used to form inferences, which subsequently effect the social judgements that shape everyday life.

Inferences and Social Judgements
Inferences play an important role in how actors understand each other and their social situations.To infer is to arrive at a judgement or conclusion based on discernable evidence.Inferences are arrived at in the mind of the observer, but informed by information provided by the observed.Inference particularly informs social interaction in settings where information is incomplete or imperfect (Llewellyn and Hindmarsh 2013).There are many examples of inference at work in everyday organisational life.An employee may call in sick but arrive back at work with a sun tan, possibly leading to the inference that she spent time at the beach.A colleague may stay late, compliment the boss and take on extra work, from which her boss may infer that she wants a promotion.Studies of the use of inference in organisational settings have focused on the requirement for one party to 'read between the lines' in order to understand the other and take appropriate action (Llewellyn and Hindmarsh 2013;Callaghan and Thompson 2002;Raghuram 2013).The abilities necessary to engage in inferential work are thought to be social and interactional rather than individual and cognitive ones since inference relies heavily on subtle cues that evade straightforward categorisation (Llewellyn and Hindmarsh 2013).For example, judgements may be formed based on age, ethnicity, and family status as well as less tangible factors such as the way another person deports themselves in terms of gesture and tone.While these characteristics are complex to decode in face-to-face interaction (Hampson and Junor 2005) the difficulties can be compounded in a technologically mediated interaction (Rettie 2009;Richey et al. 2016).In these settings information asymmetries abound, requiring all parties involved in a communication to engage in some degree of inferential work (Ravishankar 2015;Rettie 2009;Raghuram 2013).Geographically dispersed teams that may never meet each other face-to-face are obliged to form inferences based on the information made available via social technologies (Subramaniam et al. 2013;Leonardi 2014).Even those who normally work in the same physical environment can supplement their everyday social performances with information posted to their social media profiles (Leonardi 2014;Treem and Leonardi 2012).Thus, social media inferences are formed as information available via profile pages and posts is used to make judgements about a person's character and capabilities.Research on self-presentations has highlighted the central role of the audience in determining these professional selfpresentations (Goffman 1959;Schlenker and Wowra 2003;Raghuram 2013).When deciding how to present themselves, actors account for the various qualities of their audience, in particular their significance Bby virtue of their power, attractiveness, expertise and number^ (Schlenker and Wowra 2003:873).The status of the audience in relation to the actor determines the extent of their influence on the subsequent selfpresentation.In the presence of a high status audience the actor becomes increasingly concerned with giving a performance that will foster a favourable impression while annexing information that might contradict the performance out of view (Goffman 1959).In organisational settings these powerful audiences are the managers and chief executives that determine the career trajectories of those with lower professional status.These influential audience members form opinions based on the performances they observe and exert their influence over subsequent outcomes for the actors (Van Iddekinge et al. 2016;Roth et al. 2016).They may draw inferences based on a wide variety of social cues, intentionally given and unintentionally given off (Goffman 1959).Thus the actors' behaviour is noticeably adjusted to reflect appropriate and desirable social norms.These norms are shaped by powerful institutions and cultures at a societal level and can vary dramatically across contexts (Vallas and Cummins 2015;Raghuram 2013).In order to function successfully in their organisational context actors need to understand the implicit norms constituting the social order in their local setting and what part they should assume in the larger on-going performance of social life (Goffman 1959;Swidler 1986).These macro and micro cultural norms shape local attitudes about what can be appropriately disclosed by professionals both in and out of the workplace (Vaast and Kaganer 2013).
By introducing accounts about the use of social media as a source of rich, multiple inferences, our empirical sections provide insights into the preoccupation with inferences in professional self-presentation.The narratives suggest that these sources of inference can provide a valuable tool for career advancement yet are risky since they are difficult to control.Our analysis focuses on how actors cope with this tension by drawing together information from different life domains rather than drawing boundaries between them.

Methods
We adopted an interpretivist viewpoint, which Bdoes not predefine dependent and independent variables but focuses on the complexity of human sense-making as the situation emerges ( Klein and Myers 1999;Mayasandra et al. 2006).From this viewpoint, the same physical institution, artefact or human action can have different meanings for different human subjects as well as for the observing social scientist (Lee 1991).As researchers we were involved with the respondents in the process of negotiating the meaning of the data and the conclusions drawn reflect this combined effort (Corbin and Strauss 1990).
This study is based on data collected from qualitative interviews with 31 Indian IT professionals who had a graduate degree or above.In the sections that follow individuals have been given pseudonyms in order to preserve anonymity.This research setting was chosen since it is an example of a crowded and hyper competitive industry, in which it is extremely difficult for individuals to stand out and gain recognition.We focused on Indian IT professionals, who we assumed would need to engage with social media in creative ways to get ahead.Given the technological nature of their expertise we expected them to be using social media in advanced and interesting ways.Initially, we used purposive sampling techniques to identify participants that were likely to be engaging with social media for professional purposes.We then used snowball sampling techniques to encourage recommendations from participants.A call for participants was also shared via the social media platforms used by the respondents.We sought to include IT professionals from both multinational companies (MNC) and Indian IT companies.In addition to interviews, field notes were produced in the process.Table 1 shows the spread of interviews with respondents across four basic categories: Manager (greater than 10 years' experience), Program Analyst (between 7 to 10 years' experience), Associate-Projects (between 3 to 7 years' experience) and Human Resource Manager (5 years' experience).
Each interview was conducted in English and lasted roughly an hour, during which the participants responded to questions about their use of social media for professional purposes.Respondents spoke at length about their experiences of interacting on social media with colleagues, clients and other professional colleagues.They also reflected on potential consequences for their careers, respect and liking by workplace colleagues and the resulting impact on their image at the workplace.Further, respondents spoke about anxieties, fears and cognitive demands when interacting online.The need to control managerial interference was a dominant theme, which exacerbated the participants' anxiety when interacting with professional contacts on social media.
Following a qualitative, inductive method (Gioia et al. 2013) data analysis and transcription occurred simultaneously throughout the data collection period.Through this process we achieved theoretical saturation; that is, no additional themes emerged with additional data.We began our analysis by reading and summarizing each interview in order to identify recurrent themes across the data set.We identified a coding unit as a complete sentence or series of complete sentences that constituted a single semantic unit.We generated a large number of codes during this process, which were gradually reduced as we combined codes with the same meaning and excluded those that had no relevance to the research question from our list.The second column of Table 2 summarizes the codes established during this process.Having established these themes we engaged in further open coding to arrive at more abstract interpretive concepts, which are summarized in the third column of Table 2.These concepts enabled us to arrive at theoretical explanations for the harnessing of social media inferences, the potential pitfalls of using social media and the making of judgments about how to use social media appropriately.
In the findings section that follows we use the interpretive concepts as subheadings under which we group the thematic codes.The participants spoke about using social media to create an image or impression that would impress their professional colleagues and help them to stand out in the work place (CREATING IMAGES).They did this by establishing rich descriptive profile pages, posting pictures and networking with other colleagues on social media.They exerted these efforts in order to foster impressions in the minds of influential others, some of whom they had not met yet (FORMING IMPRESSIONS).The participants nurtured these impressions as the different bits of information about them on social media were pieced together to form a picture.It was not always easy to manage the impressions being formed, since the information available about the participants came from a wide variety of sources.In some instances, work colleagues became privy to information and images that were not deemed appropriate in a professional context (SOCIALLY UNACCEPTABLE).This typically occurred when individuals dropped their guard and slipped up.Negative inferences could also be made when other people made posts featuring the participants that were considered damaging to their professional reputations (DAMAGING ASSOCIATIONS).The potential for social media to both enhance and damage their professional images triggered some reflection about how to best use social media.The participants attempted to avoid damaging mistakes by engaging in cognitive reflection and imaginative techniques that served as self-checking mechanisms (SELF REGULATING).These reflective moments allowed them to audit how their social media profiles were representing them professionally.They also mentioned the role of others in regulating how elements of private life could be brought into a work context (SOCIALLY REGULATING).

Harnessing Social Media Inferences
The idea of promoting oneself online was well established for the technologically inclined IT professionals in this study.Most made use of a wide variety of blogs, forums and social media sites to endorse their expertise and professional achievements.Together these sources of information formed a professional image of who the participants were and what they were capable of.They were particularly keen to use their posts to attract the attention of influential others who could advance their career prospects.
You post online to just acknowledge that you are an expert…another reason could be that you are actually looking for other prospects and you need to advertise the fact that here you are worth something, you know this, you know this stuff, and if this is what people are looking for, then you are the person to come to.(Ankit) This was not easy to accomplish, since many of their counterparts were also using the same tactics.The result was a busy and crowded online environment in which it was tricky to get noticed.Under these circumstances the participants became more creative, drawing in additional sources of information that could add to the overall picture they presented of themselves online.These sources were not directly related to their professional lives; rather they originated in their lives outside of work.The participants used these additional sources to add interest to their professional profiles and to stand out from the crowd in positive ways.

Creating Images
The participants suggested that it was no longer possible to get noticed by expressing professional opinions and sharing knowledge since most people in the industry did this.They talked about using social media to construct images of themselves that included additional dimensions that would appeal to their employers.
Social media becomes a tool for others to understand you as a personality.I'm very conscious about that.So I use Twitter to express my opinion in a way that reflects my personality through which people understand who I am (Raghav) Most of the participants referenced the opportunity to express their personality on social media.In addition to the difficultly they had with getting noticed they also said they worried about the dominant impression that software coders were 'boring'.Social media gave them the opportunity to show additional sides of their lives.While these did not directly relate to their day-to-day work they did highlight favourable aspects of their personalities that could appeal to colleagues at work.They opened up information about their interests outside of work and how they spent their free time.They also allowed colleagues to access information about who else they were connected to and what kinds of conversations they were having.They felt that these additional sources of information reflected favourably and added to the over all impression that they were good people to work with.
The 'about me' is one section where people write elaborate things.You have your interests, you have your hobbies, your music.The kinds of things you comment on, the kind of likes that you have, the kind of posts that are generated by you, it speaks a lot about you.(Rajeev) These social media sources were not directly about work, but Babout me^and the participants recognised the potential for narcissism.They observed that on occasion their colleagues would go to great lengths to construct a social media profile that looked 'cool'.They included pictures and videos from their leisure time, they commented on current affairs and popular culture, they showcased their hobbies.Many found that this helped them build a rapport with their work colleagues and superiors.They felt that sharing this level of detail enabled them to present a well-rounded view of themselves.They were no longer just coders, but coders that also ran marathons, enjoyed photography or ran side businesses.They hoped these elements of their private lives would reflect favourably and help them to get noticed.They opened up these personal details to their work colleagues by adjusting their social media privacy settings and inviting those they worked with to link up.
You add people as connections because it's an additional glimpse of you that they can see apart from what you already told them (Aalia) They did not need to communicate everything about themselves directly in face-to-face conversations because their profiles added glimpses into their lives that inferred positive things.The participants said that they felt their co-workers attitudes towards them changed as they got to know more about them through social media.Some of them also enjoyed seeing their relationships with their bosses become slightly less formal.They felt that they gained favour with the management by allowing them to access the social media profiles that they had built.Using social media afforded them the chance to share other aspects of their lives that fostered favourable impressions in the minds of their colleagues.

Forming Impressions
The participants explained that the main reason they put effort into building a professionally appropriate social media profile was in order to foster favourable impressions in the minds of their colleagues.They recognised that they formed impressions about others by gathering additional information on social media and they assumed others would find out about them in a similar way.They said that often colleagues accessed each other's personal social media accounts out of curiosity but that these informal investigations of each other often had implications at work.By incrementally posting information about themselves for their colleagues to see they hoped to foster favourable impressions.They wanted to emphasise while they were committed to work they were not one-dimensional.To showcase their creative persona they not only posted about work related issues, but included pictures and posts suggesting that they were just as involved in their out of work interests and that they had a lot going for them.Since it was difficult for them to meet and get to know everyone in their large, geographically dispersed organisations, they often used social media to foster impressions with those they did not know well.
There are a lot of people from other locations who can't meet you so they have to form an impression of you solely through what you write or through your communicator.(Teena) The participants felt that as social media had become more ubiquitous it was more common for people to form impressions about them based solely on their online profiles.In some instances they felt that this was positive since social media could enable impression building that could not otherwise occur.They were very unlikely to physically meet colleagues in other locations but they could get to know one another through their shared associations on social media.As social media use increasingly became a feature of their working lives, they felt that they were expected to allow even those they had never met to connect with them.This made visible not only what they were posting themselves, but also what others were saying about them.They recognised that there were advantages to making use of the many different forms of inference available on the platforms.
Your work people see what other people say about you.So to that extent its more truthful than a resume would turn out to be.(Aalia) While they considered it useful to be associated with positive posts and comments from others, there was also some uneasiness.They recognised that they were not entirely in control of the information associated with their profiles on social media.

Socially Unacceptable
While the participants hoped to gain advantages by sharing their personal lives on social media, they also recognised that there were disadvantages to being so open.For example, it became challenging to anonymise their remarks, which meant they had to be increasingly cautious about what they talked about in their posts.
One of my colleagues went on Facebook and said something derogatory about salary raises without mentioning the company name.But everyone knows who you work for so these things are not accepted.(Nick) The participants talked about how oppressive it could sometimes be when everyone knew the details of your life.It became difficult to slip out of the office, to make excuses for behaviour or to share honest opinions.Although they had previously been accustomed to using social media to vent their frustrations or to be more candid in their comments, since they had begun to use their personal social media for professional purposes this was more difficult to do.It was not easy to prevent their bosses from seeing what they had posted.It was simply not socially acceptable to make negative remarks relating to work.The participants sensed that senior management checked their social media posts with the intention of reprimanding them for inappropriate behaviour.
I was called in for voicing my opinion against one of the polices that was hitting my training batch mate and myself.We were called directly and asked Byou don't want to lose your job do you?^I got really scared and after that I changed the way I wrote my posts.(Gagan) The participants were uncomfortable with the thought that social media could be used as a form of surveillance.The implications of their bosses being privy to some of their comments or practices worried them.In a highly competitive industry they knew that if they slipped up they could easily be replaced.While they were confident that what they posted to their professional blogs and work facing sites was appropriate, they were less certain about the personal social media posts that they had given their colleagues access to.The fear of being caught out on social media was intensified by the fact that others could post things about them.They spent a lot of time considering how they might avoid offending or upsetting their superiors.
I went on vacation recently and was very scared to take leave from my boss.Even though they were able to adjust their social media settings and privacy controls, there was still a high level of anxiety that with the fluid connectedness of social media, they could miss something that would damage the selfpresentations they were trying to promote.They recognised that there were a number of different ways in which their associations with different people and information on social media could be damaging to their reputations.

Damaging Associations
The participants did not feel that their social media connections would intentionally sabotage their professional impression management efforts.It was the small details that could be pieced together to form inferences that they were most concerned about.In an out of work context it would not be damaging for friends to share pictures of silly or relaxed behaviour, but if these pictures became visible to a professional contact via social media, this could potentially contradict the impression that the participants wished to foster.
If your manager had a particular view about you, you wouldn't want that to change because of one silly picture.It might not be on your profile, it could be on somebody else's.(Vasavi) They emphasised that it would not take much to damage the perceptions others had of them.It was difficult for them to anticipate the diverse ways that such damage might occur.
Many of them considered different potential scenarios in which a picture or a comment might suggest something unfavourable to their manager.They felt that there was a possibility that their managers could make connections between what was happening at work and what they saw happening on social media.
Supposing you have not been able to perform well for some reason for the past 2-3 months or a quarter.So if he's like narrow minded and not broadminded, he would link that to saying you are out too much your mind is deviating, you are not concentrating.I see judgements being passed based on Facebook.Everything is indirectly linked to there.(Vasavi) They recognised that many people formed judgements based on information from social media.They felt that the problem with this was that social media could not offer the necessary context for forming correct judgements.Furthermore, the attributions of cause based on social media posts could have profound implications for peoples' professional lives, however superficial these associations might be.
If you were to post a picture of yourself with a cigarette and a drink in your hand or acting silly, people are going to draw a lot of inferences and thereby when they see you at work they are not going to take you seriously.(Nick) The pictures and posts made visible to professional colleagues had to fall in line with the expected norms of behaviour if the participants were to be taken seriously.They felt that while they wanted to show a more interesting side of themselves on social media, there were limits to what they should display.They could not allow their on line profiles to go beyond the bounds of what was considered to be appropriate.In their particular cultural context, to be associated with smoking, drinking and partying was considered particularly unappealing.They recognised this and attempted to annex such behaviour to their private social lives.They gave considerable cognitive effort to repairing the reputational damage that might occur if such behaviours came to the attention of their managers.

Self-Regulating
The participants developed a sense of caution as they realised the level of complication involved in controlling professional impressions on social media.While they were still keen to build their online reputations in this way, they were less certain about how to limit the associated risks.They became less spontaneous as they accumulated social media connections at work, knowing that these people could now see details of their private lives.
Your colleagues are on Facebook, your friends are on Facebook.So you become very circumspect in what you say or do (Ganapathy) They hesitated in response to their concern over the potential pitfalls they might encounter.These pauses provided moments of necessary reflection.During these instances they made their own judgements about the information associated with them on social media.They considered how different pictures and posts might be interpreted by their associates.The sprawling, interconnected nature of social media meant that often this kind of contemplation became an extended exercise.
I think twice, thrice, four times before I tweet something until I'm completely sure (Raghav) Many of the participants recognised that their prolonged reflection was contrary to the commonly understood design of sites like Twitter.They understood that many of these social media sites were intended to involve responsive and immediate posting.Yet they found that giving themselves space to prepare became vital to retaining a sense of control over their social media images.They asked themselves many questions about the content of their posts, examining each from different moral and social perspectives.
I would say that it does help to look inside, be introspective, take a step back and consider whether the way you are interacting with people is right.(Prema) Their feeling that there were right and wrong ways to do things governed what they included in their social media self-presentations.In part they based their judgements on how well a post helped to support the pre-determined image they wished to present.If the picture or comment did not contradict the intended impression then in one sense they deemed it to be appropriate.The boundaries between what was 'right' or 'wrong' also shifted in relation to the different groups of people they were connected to.This was trickier for them to manage, since different colleagues responded favourably to different types of posts.Navigating this challenge require a great deal of imaginative effort.

Socially Regulated
The participants sensed that in their large organisations there were potentially hundreds of people 'watching' them via social media.They saw this as simultaneously a source of pressure and great opportunity.They harnessed this idea as a cognitive tool for coping with the social pressure of interacting on social media.Many said they imagined different potential audiences as a way of 'road-testing' their posts.
So imagine you are sitting in front of your audience, and everyone knew you personally and then say what it is you wanted to say.(Samarth) The act of imagining their social media audience enabled the participants to anticipate how their posts would be received.This technique rendered the process of making social media posts less abstract.They understood how to interact with managers and work colleagues in face-to-face settings and their acts of imagination helped them to apply the same rules to their social media interactions.On occasion some participants found that social media altered the dynamics of their relationships with their managers allowing them to engage in less formal conversations.However, on the whole they realised that they had to observe the same social norms and rules as existed in the real world.They said they came to this understanding as they were corrected by and corrected other people's social media behaviour.
The answer is to teach people to police themselves.
That's what happens on blogs where bloggers tell each other what is alright.Even on the internal bulletin board it's self-policing otherwise it won't work.You are not in any random network, this could define your career so act responsibly.Most people do, and if they do not they are pulled up and they learn.(Teena) As colleagues helped to regulate each-others behaviour, the participants felt that there was some understanding that there could be mistakes.Minor errors were tolerated but people were expected to learn and adjust their behaviour.More serious mistakes could be career defining.These involved major breaches of the established social conventions within an organisation.
During their periods of reflection, the participants kept these implicit rules in mind, reminding themselves that posts about 'bunking off' work or engaging in activity contrary to the values of their organisation could cost them their jobs.
Always keep in mind that even when you don't think about it the company's image is at stake one way or another, which is fair.If you keep in mind that even if you are not talking about them directly, one way or another you are representing the firm.You should use your good judgement and then you wouldn't do certain things.(Teena) During their periods of contemplation they drew upon their understandings of organisational values.Some recognised that their social media self-presentations should complement the image of their organisations or at least not contradict it.Their organisations became significant entities in their own rights as the participants imagined various social values with which they needed to comply.By imaginatively engaging with social actors and values they felt better prepared to anticipate the implications of each social media post and to retain control of the impressions created by their social media posts.

Discussion
Our study shows how individuals make use of social media to generate favourable inferences that help them stand out at work.While many studies have adopted a sociological perspective in describing the measures taken to delimit professional and personal social media accounts (French and Read 2013;Marwick 2011;Stutzman and Hartzog 2012;Treem and Leonardi 2012) only recently has the intentional blurring of these boundaries been acknowledged (Neeley and Leonardi 2016).These developments suggest a growing level of sophistication and savviness on the part of users who aim to benefit by using social media in creative ways.Our findings complement these recent studies and add additional nuance by reconceptualising social media performances and elaborating upon the cognitive underpinnings that support them.Our findings both engage with the well-worn metaphor of social performances and raise new questions about the type of performance social media affords.Several studies have applied dramaturgical vocabulary to social media use, debating whether these performances take place on a formal front stage, a relaxed back stage or simultaneously both?(Turner and Reinsch 2010;Marwick 2011;Hogan 2010).Questions about how actors separate different audiences and cognitively deal with conflicting social norms have dominated early understandings about how social media are reshaping organisational communications (Subramaniam et al. 2013;Leonardi and Vaast 2016).Difficulty in applying the dramaturgical metaphor has led some to question whether these are performances at all, or whether they more closely resemble other forms of expression, such as art galleries (Hogan 2010).Although participants are clearly engaged in social performances of the self on social media, there could be some limits to the applicability of Goffman's (1959) concepts, since his ideas rely heavily on the notion of a boundaried space.His conceptualisation is based on the format of classical theatre, which requires audiences and actors to each play their part in suspending belief and engaging in an alternate reality.In order to achieve this state the mechanics supporting the performance and any contradictory or distracting material is removed from view.This treatment of self-presentations is increasingly difficult to apply in the age of social media when the boundaries between different styles of performance are more difficult to identify and control (Richey et al. 2016;Ollier-Malaterre et al. 2013).Under these circumstances it can be challenging for both actors and audiences to know what kind of performance is taking place and what their response should be (Vaast 2007;Ollier-Malaterre et al. 2013).Elaborating upon our findings, we suggest that the challenges in applying this metaphor may in part be attributed to our narrow view of what constitutes theatre.Rather than discarding the performance metaphor, we propose that the inclusion of additional theatrical forms to the theoretical framework can refresh and supplement our understanding of social media's role in self presentations.
Based on our data we propose an alternative view of social media presentations as post-modern performances, akin to forms of theatre that blur the boundaries between the front and backstage to remind the audience that a performance is merely a representation of reality, not reality itself.Similarly, our participants brought content that would typically be consigned to life outside of their organisations into their social media performances.By including their personal interests and experiences in their professional performances they reminded their colleagues that there was much more to them than could be seen at work.As one participant quipped BI posted an old photo of me with a ponytail and our junior team mates came and said BI really didn't know you were that much fun^.So it's all about helping people to see the bigger picture of who you really are ( Gagan).Many examples of this style of performance can be found on the social media accounts of public figures.For example, former President of the United States, Barrack Obama includes informal photographs and family pictures on his official Twitter account, along with the description BDad, husband, President, citizen^.These boundary spanning performances are nicely articulated in the work of Bertolt Brecht (Brecht 1964), who employed a wide variety of techniques in order to stimulate audiences to think about how a performance related to their broader lives.In particular he would have actors drop out of character and address the audience as themselves, use explanatory placards, have stage directions read out loud and many other techniques that intentionally blurred the boundary between conventional front and back stage domains.Social media enabled our participants to enact similar styles of performance, allowing them to drop their formal performances temporarily, although like a post-modern troupe, they were still performing.This was a different style of performance, still intended for a formal audience, but designed to guide them in drawing new conclusions by revealing information normally related to the backstage.These glimpses fed into front stage selfpresentations although they were not traditionally associated with that realm.We argue that social media presentations can be conceptualised as post-modern forms of performance, in which individuals can intentionally disrupt the boundaries between front and back stage to infer meanings that may not be possible in a traditional organisational performance settings (Llewellyn and Hindmarsh 2013).
These hybrid performances were very demanding for the participants.Most notably, by disrupting the boundaries between front and back stage information, they relinquished a measure of control over what could be viewed by their formal audience.They were hyper-sensitive to the risks of this approach, since in many cases the status of the audience they were addressing was high in comparison to their own.While this kind of social media performance enabled them to communicate things by inference that they could not have done otherwise, there were also limits.They did not want what was revealed on social media to completely disrupt the on-going performance of social life in their actual work settings.For instance, many participants referred to certain types of content that could cause offense or argument with their colleagues, such as political or religious opinion and other socially sensitive material.The stakes for triggering these types of conversations at work were potentially very high.They were constantly alert to the rules and norms of the prevailing social order in their physical work settings.The popular affordances approach to social media implies that its uses are limited only by users' imaginative interpretations and the capacities of the technology (Treem and Leonardi 2012;Leonardi 2014).Our study draws attention back to the still powerful real world context in which posts are viewed.While social media may allow users to disrupt the boundaries between front and backstage performances, in these settings the social order that regulates behaviour retains primacy.
Concerted mental effort was required lest the participants forget the governing rules of the social order and become carried away in what they shared (Richey et al. 2016).As both formal and informal material become part of the ongoing performance of self, the private space for unchecked selfexpression is squeezed.Since they perceive these performances to be high stake activities, participants can experience intense pressure.Interestingly, rather than withdrawing their personal posts and resuming strictly separated styles of behaviour, our participants coped by creating a third cognitive space that was neither a formal front stage nor a completely relaxed back stage.The participants imagined the potential audience for their posts or forced themselves to pause and think two or three times about the content they were about to share.Many of them used the term 'introspection' to describe the time and cognitive space they needed for their social media performances.This was more like a liminal mental state, in which the actors prepared themselves for their performances.Akin to mental warm-ups, this liminal state allows actors to imaginatively address their audiences, test material and consider its appropriateness.The creation of a liminal preparatory space is crucial in enabling performers to cope with a boundary-less performance space.Our data captures the fleeting moment alluded to by Goffman, when an actor transitions from a back stage to a front stage state.Recent studies have highlighted the potential challenges that arise as different social domains collide on social media (Richey et al. 2016;Neeley and Leonardi 2016;Ollier-Malaterre et al. 2013).These difficulties have been associated with the polarisation of formal and informal contexts and social media posts.In light of our findings we propose that in a social media age, these categories are becoming less absolute; aided by social media technologies they can be actively unsettled by actors and organisations (Neeley and Leonardi 2016).Indeed, it would appear that actors are not necessarily always compelled to disclose their personal lives, but instead are opting to become increasingly transparent as they engage in these post-modern performances of the self.We propose that the mastery of a transitional liminal state is key to coping with the increasing openness demanded by organisations and practiced by individuals.

Concluding Remarks
In conclusion, this paper complements studies that assume individuals separate their different life domains on social media (Marwick 2011;Hogan 2010).The findings suggest an increased willingness to voluntarily self-disclose in a hybrid style of performance conceptualised above as a type of post-modern performance.This shift in performance style generates a number of interesting avenues for further investigation.Future research could focus on questions pertaining to 'performers' engaged in an increasingly boundary-less space (Richey et al. 2016;Neeley and Leonardi 2016;Ollier-Malaterre et al. 2013).When presentations are no longer conceived of or sustained by multi-member teams, investigations might focus on how "one-man teams" sustain and appraise the effectiveness of their own performances.They might also introduce insights into how postmodern performances are received by social media audiences.
This paper also contributes to the discussion about social media self-presentations by introducing the importance of a liminal mental state in coping with the increased levels of self-disclosure associated with these types of performances.Professional self-presentations are becoming increasingly demanding in a social media age that has embraced the 'ideology of openness' (Eisenberg and Witten 1987) as the default position.The converging pressures of an increasingly competitive job market along with the boundary blurring presentation space afforded by social media are shaping organisational norms related to self-disclosure (Chen and Sharma 2013).Our data suggests that while social media are being embraced in practice as a rich source of inference, individuals can also maintain a disciplined approach to these performances by engaging with a liminal mental state to help them to prepare for their social media performances.While our participants used these cognitive, socially informed mechanisms to prevent them from crossing a perceived line and disclosing too much, recent examples suggest that actors could intentionally reveal too much.For example, President Trump, in reference to his controversial comments on Twitter said; Bwithout the Tweets I wouldn't be here^(Financial Times 2017).Such an example suggests that while this study acknowledges the importance of the social order in establishing 'a line' of appropriate social media disclosure, future studies might fruitfully investigate professional choices to intentionally cross the line.
From a managerial perspective, it is interesting to note that individuals appear to be engaging in this type of hybrid performance, and associated self-disclosure, voluntarily.Studies suggest that the inclusion of personal information and informal interaction in a professional setting can enhance knowledge sharing and have other organisational level benefits (Neeley and Leonardi 2016;Leonardi 2014).There is little evidence that organisations are adequately tapping into or stimulating this type of social performance of self.Such performances should be an area of burgeoning interest to managers.
Our data was collected in an intensely competitive job market, where individuals were compelled to look for creative ways to stand out by drawing in a rich array of inferences on social media.We acknowledge that in less competitive settings, the same degree of self-disclosure may not be necessary.Indeed, a handful of our participants were still able to opt-out of these types of hybrid self-presentations.However, we would argue that in a global economy markets are forecast to become more, not less competitive and we anticipate that workers in many industries will find it necessary to do creative things to stand out.Also, since our data is not longitudinal, it is not possible for us to judge the extent to which social media behaviours become more creative and riskier over time.The degree to which social media self-disclosure norms drive increasingly risky and creative self-presentations is an interesting avenue for future studies.

Table 1
Number of interviews across role categories

Table 2
Coding structurePeople use these exaggerated tales to convey an image of being cool.(Sylvia).CREATING IMAGES Harnessing social media inferences There are a lot of people from other locations who can't meet you right so they have to form an impression of you solely through what you write or through your communicator or whatever.(Teena) You find someone at the office and you get to know them on social media.Because once they see more personal stuff on Facebook they like it, you start to get closer, they start appreciating what you do.It acts as an icebreaker.(Ganesh)