August 28, 2020

In-Depth: What a Social Systems Perspective Teaches us About Change

Banner

In this series of ‘In-depth’-posts, we take time to consider the bigger picture. This series is not about easy answers or practical tips, but to develop a more complete understanding of what may be going on. You can also listen to a recording if you prefer to listen instead of read.

If you take the time to fully read and digest this post, you will (hopefully):

  • Understand how a social systems perspective gives you a better way to understand change, how it happens and what you can do to facilitate it.
  • Recognize how much of your thinking is still subtly influenced by a mechanical perspective in organizations .
  • Be familiar with insights from five areas of scientific research that explore how the performance of teams and individuals may depend more on the situation — like network patterning, mood, safety, social contact and social influence — than on individual traits.
  • Realize just how deep the “rabbit hole” is when you apply insights from this posts in your work, in teams, to our society and to yourself.

The writing and research for this post was made possible by our patrons. If you like it, and you’d like more of it, please consider supporting us too.

Why Is Change Hard?

When organization or teams change how they work — like choosing for Scrum — they effectively introduce new norms and behaviors into their group and the people they work with. These norms range from very basic ones, like attending a Daily Scrum every day, to abstract ones like “work together towards a shared goal instead of working on your own tasks” and “delivery small, working bits of product frequently instead of huge chunks occasionally”.

We all know how changes can seem so simple, but turn out to be annoyingly hard. Why is that? If you find yourself jumping to explanations like “resistance”, “people don’t like new things” or “older people prefer to stick to their ways”, you may have fallen victim to one of the limitations of the mechanical perspective on organizations.

Organizations are not machines. Although this may seem obvious, we often still understand organizations, and how the people in them work, from this perspective. The idea of the organization as a machine has become so ingrained and pervasive in management practices and theories in the past fifty years, that it is hard to even be aware of it.

The mechanical perspective certainly has its uses. At the same time, its dominance easily leads to huge blind spots in our understanding of what organizations are and how the people in them do their work. In this post we explore one its largest blind spots: change, or why it happens, how it happens and what can facilitate it.

The Organization as a Machine

The mechanical perspective draws attention to the structure and processes of organizations. The organization is seen as a structure that can be broken down into ever smaller, discrete units. From “organization” to “business units”, “departments”, “restructuring”, “transformation”, “work-groups”, “teams”, “roles” all the way down to individual tasks. The mechanical perspective puts a strong emphasis on efficiency, saving time, and optimizing flow between these units. It shows in words we often still use, like “human resources”, “change implementation”, “job descriptions”, “escalating to the line” and “work packages”.

Morgan (2006) explores the roots of this mechanical perspective — or the “machine metaphor” — and attributes much of it to scientific management. This approach to management was developed by Frederick Taylor between the 1880s and the 1900s and was firmly based on economical efficiency. This is apparent in the four principles of scientific management: 1) analyze, measure and optimize individual tasks, 2) divide work functionally so that managers plan the work and workers execute it, 3) select, hire and train people for their tasks and 4) management cooperates with workers to ensure that scientifically developed practices are followed (Taylor, 1911). 

Although scientific management was developed primarily to optimize repetitive work in factories, it quickly spread to other domains. In the following century, it would give rise to modern-day practices like lean manufacturing, Six Sigma, and Gantt-charts. The strong focus on empiricism and measurement would later inspire Kanban and the Scrum Framework.

The most significant criticism of scientific management — and the mechanical perspective behind it— is the total lack of consideration for human needs and interactions. Morgan (2006) draws attention to its deeply reductionist nature. Instead of seeing people as unique individuals, they are reduced to a collection of skills that are mapped to standardized job descriptions and then trained to become better at them. When teams are formed, this is done with a checklist of skills in mind. There is no consideration for the unique dynamic of potential members, the needs they bring, and the beliefs they have. In the end, the internal logic of the mechanical perspective rests on checklists, managerial plans, rules, processes, and numbers.

By taking human interactions out of the picture, the mechanical perspective paints a very incomplete picture of what organizations are and which interventions are helpful. It also doesn’t help us understand why selecting the best people doesn’t automatically result in a stellar performance. Or why organizational change initiatives are often met with “resistance”. 

In this post, we consider a perspective that begins with human needs and interactions instead. It is based on decades of work by sociologists, psychologists, anthropologists, and management researchers and helps us understand more about what is happening in organizations on a human level.

The Organization as a Social System

In the social systems perspective, organizations are considered as loose groups of people that interact more closely with each other than with people outside their group. Although these loose groups may overlap with the departments and names on an organizational chart to some extent, their boundaries are not as clear as the diagram suggests. Although people from the sales department may interact more closely with people from their own department, they also interact with others at the water-cooler, in meetings, during lunch, or for drinks after work. 

Where do you draw boundaries when an organization is more like this? Picture by Gordon Johnson from Pixabay.

There is no clear boundary where the organization begins and ends either. From a mechanical perspective, the question of what is and isn’t the organization is often answered with an organizational chart, the address of its buildings, or the roster of employees. From a social systems perspective, the answer would include the people that frequently interact with groups “within” the organization — like temp workers, customers, former employees, spouses, and families of employees, even competitors. It wouldn’t be limited to the present and include the historical accumulation of social norms and beliefs about “how we do things here”, usually referred to as “organizational culture”.

This may seem like a strange way to think about organizations if you’ve never considered otherwise. A similar comparison can be made for how we think about countries. We could consider The Netherlands — where I live — as the accumulation of everyone who currently lives within its official borders. This is similar to a mechanical perspective. What about people who migrated, but still strongly identify as being from the Netherlands? What about people living along the borders of our countries that may just as well describe themselves as Germans or Belgians, or a mix, or don’t care at all? What about the different groups that have lived in the Netherlands historically, bringing and evolving their customs, social norms, and beliefs? Considering a country as only the people living there right now, or those that have a passport is an oversimplification.

The social systems perspective on organizations makes things muddier and more complex. Contrary to the mechanical perspective that reduces work to clean and rational tasks, roles, processes, and structures, the social systems perspective starts from our intrinsic social nature. Although our individual preferences for socialization vary, our ancient ancestors found safety, warmth, and food in groups. Being, and remaining, part of a social group has always been an important survival strategy for our species. Even though it may not be as critical today as it was, we’re still deeply attuned to what is happening in groups. Who is the ‘leader’? How do others behave? What do others think of me? Who likes whom, and who doesn’t? Which groups are allies? Which are enemies?

Different Perspectives on Change

There is much to say about the differences between these perspectives (see also Morgan, 2006). For the purpose of this post, I focus on how they approach and understand “change”.

From the mechanical perspective, change is defined in terms of modifications to structures and processes. Consistent with the principles of scientific management, these modifications are identified, designed, and implemented by management — often with support from external consultants. Employees are then expected to follow, and any reluctance to do so is interpreted as resistance. A good example of this is when employees are handed new “task procedures” for the work they’ve always done. Or when job descriptions are changed throughout the organization to fit the desired goal without regarding the employees currently performing those roles. From a mechanical perspective, the organization is changed during a specific moment in time and people are expected to follow.

“In the mechanical perspective, the organization is changed during a specific moment in time and people are expected to follow.”

The social systems perspective considers change as a social process where the rules, norms, and beliefs of people shift due to social influence and group dynamics. In this perspective, how people behave in an organization is shaped by their social environments. If most colleagues act dismissive to an angry customer, everyone will soon start acting that way. When someone with a high social status acts in a deeply unethical manner (e.g. Groupthink and what we’ve seen at UBER), others will soon follow. This happens regardless of structures, value manifestos, and procedures — social influence is that much stronger. Although these examples reflect undesirable behavior, its good to note that the same social influence applies to desirable behavior. In the social systems perspective, change happens all the time as people continuously shift their rules, norms, and beliefs in response to what others do.

“In the social systems perspective, change happens all the time as people continuously shift their rules, norms and beliefs in response to what others do”

When it comes to change, the mechanical perspective struggles to explain a lot of what we commonly observe in organizations during and after the change:

  1. When employees are reluctant to follow in the change, the only real explanation that the mechanical perspective offers is that of resistance. More specifically, it assumes that people are generally unwilling to change their behaviors and beliefs due to personal traits (like age, intelligence, education).
  2. Because the mechanical perspective implicitly treats employees as unwilling to change, it has given rise to a variety of management practices that are designed to compel change from the outside. Examples of this are individual reward systems, the use of force (change or risk getting fired), and “management by objectives” (Drucker, 1954).
  3. Because the mechanical perspective treats change as something that happens only at specific moments in time, it can’t explain how — after a change — performance drops or behaviors revert back even when the structures and processes remain the same.
  4. With its reductionist approach, the mechanical perspective assumes that the performance of larger units (e.g. departments, teams) is simply the sum of its components (e.g. individuals). The mechanical perspective doesn’t help us understand how the situation shapes the interactions between these components, and how they result in higher or lower performance.

Below, I draw from five different areas of research that demonstrate how a social systems perspective helps us understand more about change.

1. Situations Shape Behavior More Than Personal Traits

From a mechanical perspective, individual employees are the smallest components that make up the organization. By selecting the brightest “components” and putting effort into their individual development, it is assumed that the system as a whole will perform better. This is why so many organizations invest heavily in recruitment based on personal traits; intelligence, personality traits, skills, and personal values. An implicit assumption here is that the behavior of individual employees is mostly determined by the personal dispositions they bring to the workplace.

The social systems perspective brings with it a different understanding of what determines behavior. Instead of drawing attention to personal traits, it considers how behavior is shaped more by the situation that someone is in than their personal traits. Although this may sound counter-intuitive, there is a lot of empirical support for this hypothesis.

For example, Ross, Nisbett & Gladwell (2011) offer an extensive overview of studies that show how personality traits only explain a limited amount (usually between 10 and 30%) of variation in our actual behavior. The remaining variation seems to be explained primarily by the situation that someone is in. In practice, this means that someone who scores high on introversion on a personality test will act in a manner that is consistent with introversion in some cases, but not in most cases when then the situation is different. In another example, researchers found that dimensions from the MBTI personality-test only explains some of the behavior of managers (Gardner & Martinko, 1996) but that the situation — like team composition, social norms, organization size, communication patterns — determine most of it. Despite the popularity of the MBTI and similar personality tests, the empirical evidence for the ability of these tests to predict actual future behavior is very limited (Pittinger, 2005).

This research has deep ramifications for how we understand what motivates our behavior and that of others (Ross, Nisbett & Gladwell, 2011). For the purpose of this post, it shifts the focus from “individual skills and qualities” to “creating the right environment”.

2. Mood, Humor Patterns and Psychological Safety

As we saw, the mechanical perspective assumes that the quality of groups (departments, teams, work-groups) is simply the accumulation of individual skills, talents, intelligence, and personalities. From this perspective, the best teams are those with the cleverest, brightest, and most creative professionals. Although some lip-service may be paid to team-building, psychological safety, and conflict navigation, most effort goes to selecting the best components.

The social systems perspective draws attention to how social processes on the level of groups impact their performance regardless of what individual members bring to the table.

One example is an observational study where Lehmann-Willenbrock & Allen (2014) observed 54 organizational meetings of real teams. They found that “at the team level, humor patterns (but not humor or laughter alone) positively related to team performance, both immediately and 2 years later”. In another example of how situations shape behavior, they also found that the positive influence of humor disappeared in teams where people worried about job security (e.g. looming layoffs and reorganizations).

Not only humor but mood, in general, is conducive to performance. Grawitch, Munz & Kramer (2003) demonstrated experimentally that groups with positive moods tend to generate more original ideas than groups in a negative mood (12% of the variance was explained by mood alone). Although mood is initially something we individually bring to the workplace, it is susceptible to a process of social contagion where it spreads to others (Le Bon, 1895). As another example of our social nature and how we look to others, we tend to copy moods, emotions, and behaviors more from people we like or hold in high regard (ibid).

Another example is psychological safety, often defined as “as a shared belief about the consequences of interpersonal risk-taking” (Edmondson, 1999). In a study with 51 teams, she demonstrated how psychological safety strongly contributes to team learning behavior, which in turn influenced performance. Psychological safety accounted for 33% of the variance in observed learning behavior and 63% of the variance in self-reported learning behavior of the team. Interestingly, Edmondson (ibid) also found that a shared belief about a team’s capabilities did not significantly contribute to team learning behavior — psychological safety alone was enough. A similar result was found in a study at Google of more than 180 teams over a period of 2 years (reported in Duhigg, 2016). It concluded that psychological safety was the strongest contributor to team performance. Our own research with 1.195 Scrum Teams, showed a similar pattern. We found that psychological safety was nearly indistinguishable from our operationalization of “cross-functionality” and was one of the strongest contributors for team morale, release frequency, and focusing on value (publication will follow soon).

In our own research, cross-functionality turned out to be one of the strongest predictors of team morale, team value, and stakeholder satisfaction (through value focus and release frequency).

It is important to underscore that what is happening in these studies on the level of teams is not simply the accumulation of personal traits and beliefs. As Edmondson (1999) writes about psychological safety: “The data […] suggest that team psychological safety is something beyond interpersonal trust; there was evidence of a coherent interpersonal climate within each group characterized by the absence or presence of a blend of trust, respect for each other’s competence, and caring about each other as people”. It’s a compelling thought; instead of the people making the team, the team seems to make the people. We’ll explore more evidence for this hypothesis later in this post.

“Instead of the people making the team, the team seems to make the people”

3. Social Interaction and Performance

From a mechanical perspective, social interaction is often seen as a distraction from work. A good example of this is field research that was performed in a call center in 2008 (Pentland, 2012). For a while, the researchers followed four teams of twenty employees each. Because the work was highly standardized, with prepared call-scripts and detailed call-time metrics, office policy was designed to reduce social interaction to a minimum. One example of this was how the breaks for coffee and lunch for teams were staggered, instead of taken together. The reasoning was that the work was so standardized that employees wouldn’t benefit from the experience and help from colleagues. The researchers suggested changing the policy to allow teams to have their breaks together and to give people more opportunities to socialize, away from their desks. Pentland (ibid) notes that employee satisfaction increased, sometimes up to 10%. Although already beneficial on its own, the researchers discovered that average call-time “fell by more than 20% among lower-performing teams and decreased 8% overall”. In terms of productivity, this represented an approximated increase of 15 million dollars a year. A change that didn’t make sense from a purely mechanical perspective, turned out to boost both satisfaction and productivity.

By spending their coffee breaks together, call-center employees were able to learn from each other and perform better (Pentland, 2012). Picture by Free-Photos from Pixabay.

4. Change, resistance, and social influence

How do people, teams, and organizations respond to change? From a mechanical perspective, the hesitation of people to try something new is often interpreted as “resistance to change”. In many cases, those “driving the change” — another implicit machine metaphor — are quick to attribute it to personal attributes, like age, education, intelligence, and personality.

A social systems perspective paints a richer picture of what happens here. From this perspective, change (any change) is effectively a social process that involves modifying social norms, habits, and “the way things are done here”. Because we are social creatures, we look to others to decide how to act ourselves. If a person that we hold in high regard considers a “Daily Standup” as awkward, most of us will follow suit in adopting this belief. If most people in our vicinity continue to act the same way as before, we are unlikely to change our behaviors. Social influence and group dynamics tend to be stronger than any rational argument you can make (see also Lewin, 1948, 1951).

The goods news is that the social systems perspective also offers different strategies for change that take our social nature into account. Ross, Nisbett & Gladwell (2011) cite numerous examples of how social influence and group dynamics can change behaviors dramatically. For example, they share a unique experiment by Coch & French that involved 600 workers in a factory. Faced with a series of sensible changes in production methods, the researchers divided the workers into three groups. The first group was simply informed of the changes and expected to follow them. The second group selected a few representatives that worked with management to understand the need for change, and then help their colleagues implement them. In the third group, all workers were treated as equal representatives and everyone helped each other implement the changes. The “representation approach” of the second group and the “total representation approach” of the third group resulted in no loss of morale or complaints, whereas the first group saw 17% of employees leave the company. In terms of productivity, the second and third groups reached regular levels of productivity within days compared to more than eight weeks for the first group. For the third group, productivity actually increased by 15% afterward.

Of interest to us here is that while the change in production methods could be seen merely as a mechanical change where employees switch from one set of instructions to another, it was in effect a social change where people had to change habits. When this change was forced onto workers, it created resistance. But when an effort was made to involve workers, and they saw their peers participating in it, resistance evaporated.

5. The Network Makes The People

From a mechanical perspective, we often assume that “the people make the team”. So there is a natural emphasis on hiring the best and brightest, selecting on and developing personal skills, and rewarding personal achievements. A social systems perspective, on the other hand, turns this upside down by stating that it is more likely that the “network makes the people”, as Tasselli, Kilduff & Menges (2015) summarize. In their review of research done in this area, they provide many examples that illustrate how identities, beliefs, and behavior are deeply influenced by the social networks we are a part of. For example, people that belong to the same group often start displaying the same behaviors, same beliefs and same attitudes over time — regardless of their personal dispositions when they enter (e.g. Krackhardt, 1999). Our social nature makes us flexible in how we adjust our beliefs and behaviors in order to join and stay in a social group that matters to us. Just look at how teenagers are shaped by the groups they join in high-school, the friends they have, and the people they meet through hobbies and interests. Although the shifts may be less dramatic when people grow older, this social flexibility remains (though more so in some than others). Powerful examples of this are to be found, among other areas, in research on social comparison theory and cognitive dissonance (see Festinger, 1954, 1957).

The position we take in a network shapes our performance. Image by OpenClipArt-Vectors from Pixabay.

But aside from our own social desire to belong to groups, networks also influence our beliefs and behaviors in other ways. For example, Krause, James & Croft (2010) explored how our position within a network influences our behavioral strategies — something they call the “network personality”. Burt (2012) used data from virtual groups to explore how the position in the network influences performance and concluded that “The dominant factors predicting achievement in a role are a person’s experience in the role and the network advantage built up in that role”. This doesn’t mean that personal traits have no influence, just that it is a small piece of a larger puzzle. A further demonstration of how the structure of the network shapes performance is offered by Pentland (2012, 2015). By collecting real-time data from sociometric badges and e-mail patterns, he and his team were able to demonstrate that groups perform better, generate more creative ideas, and show higher levels of engagement when the interactions within groups are increasingly equalized to include everyone (rather than just a few dominant voices).

So What?

In this post, I shared research from five domains that illustrate, in one form or another, how we are part of, and influenced by, social systems. This may seem obvious but often clashes with common practices. For example:

  • When employees are compared or rewarded based on their individual performance, completely disregarding the social environments they operate in.
  • When senior leaders (like CEO’s) are awarded huge bonuses for the success of their organization because it is assumed that their personal traits or behavior made this success possible.
  • When candidates for jobs have to participate in all sorts of individual assessments, ranging from IQ-tests to personality tests (MBTI, DISC, Big Five, OCEAN) to determine if they’re a good fit. Note: “fit” as a word is a subtle example of the mechanical perspective at work. 
  • When the composition of teams is (frequently) changed at will by outside parties (e.g. HR) without the clear involvement of the team.
  • When people offer well-meaning platitudes like “hiring the best people” or “getting the right people in the right place”.
  • When organizational change essentially boils down to management presenting the plans, and expecting employees to follow (without further collaboration).
  • When organizations sidestep team-building of any kind in favor of getting the work started as soon as possible. Instead of considering team-building as foundational to the success of a team, it is considered a “nice to have” that can be done when there are time and money left.

How Can a Social Systems Perspective Help Us?

A social systems perspective gives us a different, and arguably more complete lense, to understand what change is and how it can be facilitated. For example:

  • It is easier and more convincing to do something when peers are doing it too. So when you need to change something, change it together. The more people are involved, the more likely it is that others will follow.
  • Our social nature means that emotions and beliefs spread throughout the social network. When a few people complain about something, everyone soon starts complaining about that thing. The reverse is also true; model different behavior with a few people and see how it spreads.
  • When it comes to teams, how people interact puts more weight on the scale than the individual qualities they bring to the table. So work hard to interact in a way that allows every voice to be included. Liberating Structures are a huge help here. Spend time together to build safety between people.
  • Resistance is a social process, not a trait that is inherent to people. Although individually, some may be more conservative than others, the social desire to follow peers is likely to override this hesitation.
  • Instead of trying to “hire the best people”, work hard to create psychological safe, motivating environments where everyone feels that their voice matters. Teams make people more than people make teams. Don’t hire toxic employees, no matter how good they are.
  • Our social nature means that we are (highly) susceptible to manipulation through social influence, to the point of transgressing ethical norms. Groupthink is a good example of this, but so are cults, sects, and radical groups. So is a culture of machismo (like at UBER) where men socially encourage each other to treat women in highly sexist ways.
“Teams make people more than people make teams.”

Closing words

In this post, I contrasted the still-dominant mechanical perspective on organizations with a perspective based on social systems. Where the former concerns itself with structures, processes, and efficiency, the latter is all about people and their interactions. The social systems perspective underscores that behavior may be shaped more by the situation than the person.

Although I limited this post to the work environment, the ramifications of the social systems perspective reverberate through society at large. It has much to say about privilege and power, and how our personal opportunities are often shaped more by our environments than anything else. This message is hopeful in the sense that changing environments creates new opportunities. In another sense, the dominance of the mechanical perspective shows how much work there is still to do to.

Writing this post and gathering relevant research took about 26 hours. We’d love to write more content like this. If you find this work valuable, please consider supporting us so we can find more time (and worry less about bills). Click here to learn more about how you can support our work

References

  • Burt, R. S. (2012). Network-Related Personality and the Agency Question: Multirole Evidence from a Virtual World. American Journal of Sociology, 118(3). p.p. 543.
  • Duhigg, C. (2016). What Google Learned From Its Quest To Build The Perfect Team. The New York Times Magazine, Feb 28, 2016, pp. 20. Retrieved on June 30, 2020, from https://www.nytimes.com/2016/02/28/magazine/what-google-learned-from-its-quest-to-build-the-perfect-team.html
  • Drucker, P., The Practice of Management, Harper, New York, 1954; Heinemann, London, 1955; 2nd edition, Butterworth-Heinemann, 2007.
  • Gardner, W. L. & Martink, M. J. (1996). Using the Meyers-Briggs Type Indicator to Study Managers: A Literature Review and Research Agenda. Journal of Management, 22(1), pp. 45–83.
  • Grawitch, M. J., Munz, D. C., & Kramer, T. J. (2003). Effects of member mood states on creative performance in temporary workgroups. Group Dynamics Theory Research and Practice, 7(1), pp. 41–54.
  • Krackhardt, D. 1999. The ties that torture: Simmelian tie analysis in organizations. In S. B. Bacharach, S. B. Andrews, & D. Knoke (Eds.), Research in the sociology of organizations: Vol. 16. Networks in and around organizations: 183–210. Stamford, CT: JAI Press.
  • Krause, J., James, R. & Croft, D. P. (2010). Personality in the context of social networks. Philos Trans R Soc Lond B Biol Sci. 365(1560). P.p. 4099–4106.
  • Lehmann-Willenbrock, N., & Allen, J. A. (2014). How fun are your meetings? Investigating the relationship between humor patterns in team interactions and team performance. Journal of Applied Psychology, 99(6), 1278–1287. In Rogelberg, S. G.(2019), The Surprising Science of Meetings: How You Can Lead Your Team to Peak Performance. Oxford University Press. ISBN: 0190689218.
  • Le Bon, G. (1895). Psychology of Crowds. Sparkling Books edition. Sparkling Books. Penguin Books. ISBN: 0143126334
  • Lewin, K., and Dorwin Cartwright (Ed.) (1951). Field theory in social science. New York: Harper.
  • Ross, L., Nisbett, R. E., & Gladwell, M. (2011). The Person and the Situation: Perspectives of Social Psychology, 2nd ed. Pinter & Martin Ltd. ISBN: 1905177445.
  • Lewin, K., and Gertrude W. Lewin (Ed.) (1948). Resolving social conflicts: selected papers on group dynamics (1935–1946). New York: Harper and Brothers.
  • Morgan, M. (2006). Images of Organization. Sage Publications. ISBN 1412939798.
  • Pentland, A. S. (2015). Social Physics: How Social Networks Can Make Us Smarter.
  • Pentland, A. S. (2012). The New Science of Building Great Teams. Harvard Business Review. Retrieved on June 12, 2020, from https://hbr.org/2012/04/the-new-science-of-building-great-teams
  • Pittinger, D. P. (2006). Cautionary comments regarding the Myers-Briggs Type Indicator. Consulting Psychology Journal Practice and Research, June 2005.
  • Taylor, F. W. (1911). The Principles of Scientific Management. New York: Harper.
  • Tasselli, S., Kilduff, M. & Menges, J. I. (2015). The Microfoundations of Organizational Social Networks: A Review and an Agenda for Future Research. 41(5). P.p. 1361–1387.
  • Waber, B. N., Olgiun, D. O., Kim, Taemie, & Pentland, A. S. (2010). Productivity Through Coffee Breaks: Changing Social Networks by Changing Break Structure. SSRN, Retrieved on June 12, 2020, from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1586375