How to Assemble an Allstar Team:

The Process May Surprise You

Attracting and retaining top talent consistently ranks among the top concerns for global CEOs.[1] To succeed in this increasingly competitive arena, organizations spend an incredible amount of time and energy ensuring they get the right people in the door. They attend particular networking events, post jobs in strategic places, and may source hundreds of resumes for a single open position.[2] Candidates that advance in the hiring process often face three or more interviews before an offer is extended, and that offer may kick off a lengthy negotiation process. Companies expend considerable effort to fill each vacancy with an “all-star”.

There is no doubt that all-stars can add more value than average employees, though just how much more productive they are is up for debate. One study found that, across professions, the top 20% of performers produce 50% of the output.[3] Bain & Company’s research suggests that top employees are roughly four times as productive as their average counterparts.[4] McKinsey reports that the gap between high and low performers increases with job complexity, citing on average high performers in “low complexity” jobs produce 50% more than their average peers and high performers in “high complexity” jobs produce 125% more than their averages peers. For employees performing extremely complex work, they estimate top performers are more than eight times as productive.[5] As individuals, star performers can produce staggering results.

In today’s business environment, work is increasingly collaborative and team-based, thus expecting company success based on how employees perform as individuals is misguided. And, fortunately, if we zoom out from the individual level and instead look at the difference between the best and worst teams, the gap is even larger. In Scrum: The Art of Doing Twice the Work in Half the Time, Jeff Sutherland describes a study of 3,800 diverse team-based projects, from financial analysis at an accounting firm to development at IBM. The difference in how long it would take a low-performing team to produce the same result as a high-performing team was far greater than any of the estimates of individual performance variation – what the best team could do in one week would take the worst team two thousandweeks.[6] While the difference in output between high- and low-performing individuals is significant, the difference between high- and low-performing teams is far greater. Accordingly, the appropriate place for us to focus as leaders is not in simply hiring great people, but in creating high-performing teams.

Assuming that a team of high-performing individuals automatically translates to a high-performing team is a mistake frequently made across industries. One famous example is the 2004 United States men’s Olympic basketball team. Assembled from the best National Basketball Association (NBA) players from around the United States – including rising stars and accomplished veterans – the team was destined to take home the gold medal. Yet, in their first game, they suffered a stunning upset against Puerto Rico, losing by nineteen points. Despite what was widely recognized as a group of superior players, the stars played as individuals, not as a team, and therefore were unsuccessful in competition.[7] Their story lives on more than a decade later as a cautionary tale to managers who assume they can assemble a winning team just by putting together top performers.

Knowing how to assemble a high-performing team is a competitive advantage in any industry, but it is an extraordinary challenge. In 2012, Google embarked on a quest to develop a formula for building a high-performing team. Under the code-name “Project Aristotle”, a team of organizational psychologists, engineers, and data scientists mined through all the data they could find inside and outside of Google on this topic. Surprisingly, in their research they found no correlation between the effectiveness of a team and its composition. One People Analytics manager told the New York Times, “We had lots of data, but there was nothing showing that a mix of specific personality types or skills or backgrounds made any difference. The ‘who’ part of the equation didn’t seem to matter.’’[8] Coming from a company like Google, which grew to prominence through powerful predictive analytics, this is particularly astonishing. They found that the following variables were not associated with a team’s success: individual performance of the team members, extroversion of the team members, teammate seniority, team size, and co-location of team members.[9]

Those who believe assembling a team of all-stars is the path to success should heed warning of this finding – individual performance of team members was “not significantly connected with team effectiveness” at Google.[10] Team dynamics, rather than team composition, impacted the quantitative and qualitative metrics of team effectiveness. Specifically, they found that psychological safety, dependability, structure and clarity, meaning, and the impact of the team’s work were the greatest predictors of a team’s effectiveness. This finding is consistent with a recent study, which showed that a “collective intelligence” factor explains how groups perform on a variety of tasks. Collective intelligence is not strongly related to either the average intelligence of group members or the intellect of the most intelligent group member, but instead is related to how socially sensitive the group members are, how evenly distributed their conversations are (i.e., no single individual does all the talking), and the proportion of females in the group.[11] Team dynamics play a significant role in how successful a team will be.

From our years of studying teamwork across private industry and the U.S. military, CrossLead has developed a model for achieving and maintaining a state of high performance at the team-level. This model centers on four capabilities: trust, common purpose, shared consciousness, and empowered execution. Trust and common purpose are required for any team to be high performing, regardless of context. Trust at the team-level takes two forms: competency and benevolence. Competency-based trust means that individuals on the team can be expected to complete the work they agree to and meet quality standards (Google’s project Aristotle called this “dependability”). The second form of trust, benevolence-based trust, means that team members assume positive intent in one another. This is closely related to the concept of psychological safety, a term coined by Harvard Business School Professor Amy Edmondson[12], to describe an environment that is safe for interpersonal risk taking (the factor Google found to be most important to the success of a team). At CrossLead, we believe these two forms of trust are critical to team performance.

The second team capability we believe is fundamental to team success is common purpose. Aligning goals across the team means that the team either succeeds or fails as a collective, not as individuals. In teams intentionally assembled of all-stars, egos may get in the way of achieving a common objective because individuals are not willing to subordinate their personal goals, even when doing so is in the best interest of the group. Common purpose is most effective when members of the team find their collaborative work personally meaningful and fulfilling.

These two capabilities – trust and common purpose – are critical for any team’s success. Whether the team is a group of firefighters entering a burning home, basketball players trying to win a gold medal, or software engineers working in a garage to unseat an industry software titan, the team will not succeed without trust and common purpose. But, in today’s complex environment where organizations are increasingly expected to deliver integrated solutions, collaborating well internally is insufficient. The trust and common purpose present at the team level must extend across the enterprise as well. Success in the context of a large organization requires healthy cross-team dynamics.

The two additional capabilities – shared consciousness and empowered execution – are uniquely important in the context of large organizations, where cross-team dynamics are just as important as team dynamics. Shared consciousness is an emergent intelligence that results from consistent and meaningful cross-team interactions. At a start-up or small company, shared consciousness forms naturally. When everyone is sitting at the same table and hearing the same conservations, the team will be immersed in what is happening elsewhere in the enterprise. This allows them to act in a synchronized fashion without explicitly planning to do so.

When a small company grows into a large enterprise, shared consciousness is equally important but much harder to sustain. As more structure is added to the organization, silos form and cross-functional interaction becomes rare unless it is intentionally architected. Failing to prioritize shared consciousness has several detrimental results – duplication of work across functions, silos reaching different conclusions from the same data, teams executing in strategically different directions, to name a few. If lessons learned in one part of an organization are not reaching other corners of the enterprise, teams will repeat the same mistakes. Shared consciousness enables teams and organizations to synchronize, learn, and adapt together.

Finally, the last capability required for a high-performing team is empowered execution. Teams are most effective when they are empowered to solve the problems that are closest to them. While it is important to provide guidance to teams on what outcomes are expected, how they achieve those outcomes should be up to the team. Often managers seek to grab control by demanding that teams follow a specific process for executing their work. Instead, leaders should enable the team to reach its innovative potential by giving them the freedom to operate independently, setting boundaries only where absolutely necessary. This is particularly important in the enterprise context, where those at the “management” level are often far away from the front-lines where the strategic plans are executed.

Being part of a high-performing team is not an experience reserved for the best and the brightest. The teams that get the best results are those who build and maintain trust and common purpose and exist in organizations with free information flow and empowering structures. These are the characteristics of a true “all-star team” – a team whose superior dynamics enable them to be far greater than the sum of their parts. Teams that lack these dynamics will fail, no matter how many “all-star” players are on them.

By David Silverman

This article was published in collaboration with the Indian Management Journal.

[1] Lublin, Joann and Fuhrmans, Vanessa. “Hackers and a shrinking talent pool, top CEO concerns.” Wall Street Journal. January 3, 2018.

[2] Cherian, Josh. “50 HR & Recruiting Stats that make you think.” January 20, 2015. Glassdoor.

[3] Augustine, N. R. “Augustine’s Laws and Major System Development Programs.” 1979. Defense Systems Management Review: 50-76.

[4] Mankins, Michael; Bird, Alan; Root, James. “Making Star Teams out of Star Players.” Harvard Business Review. January-February 2013 issue.

[5] Keller, Scott and Meaney, Mary. “Attracting and retaining the right talent.” McKinsey & Company. November 2017.

[6] Sutherland, Jeff. “Scrum: The Art of Doing Twice the Work in Half the Time”. Pages 42-43. September 30, 2014.

[7] Limpert, Caroline. “Lessons from Barketball’s Olympic Dream Team”. July 11, 2012.

[8] Duhigg, Charlies. “What Google Learned from Its Quest to Build the Perfect Team”. New York Times. February 25, 2016.

[9] “Guide: Understsand Team Effectiveness” Rework.

[10] “Guide: Understsand Team Effectiveness” Rework.

[11] Williams Woolley, Anita; Chabris, Christopher; Pendland, Alex; Hashmi, Nada; Malone, Thomas. “Evidence for a collective performance factor in the performance of human groups.” Science.

[12] Edmondson, Amy. “Building a psychologically safe workplace.” Originally presented at TEDxHGSE.