From education to employment

Project Aristotle: Discovering the secrets of effective teams at Google

Much of the work done at Google, and in many organizations, is done collaboratively by teams. The team is the molecular unit where real production happens, where innovative ideas are conceived and tested, and where employees experience most of their work. But it’s also where interpersonal issues, ill-suited skill sets, and unclear group goals can hinder productivity and cause friction.

Following the success of Google’s Project Oxygen research where the People Analytics team studied what makes a great manager, Google researchers applied a similar method to discover the secrets of effective teams at Google.

Code-named Project Aristotle – a tribute to Aristotle’s quote, “the whole is greater than the sum of its parts” (as the Google researchers believed employees can do more working together than alone) – the goal was to answer the question: “What makes a team effective at Google?”

Defining what makes a “team”

The first step in answering this question of “what makes an effective team?” is to ask “what is a team?” More than an existential thought exercise, actually figuring out the memberships, relationships, and responsibilities of individuals all working together is tough but critical to cracking team effectiveness.

The term team can take on a wide array of meanings. Many definitions and frameworks exist, depending on task interdependence, organizational status, and team tenure. At the most fundamental level, the researchers sought to distinguish a “work group” from a “team:”

  • Work groups are characterized by the least amount of interdependence. They are based on organizational or managerial hierarchy. Work groups may meet periodically to hear and share information.

  • Teams are highly interdependent – they plan work, solve problems, make decisions, and review progress in service of a specific project. Team members need one another to get work done.

Organizational charts only tell part of the story, so the Google research team focused on groups with truly interdependent working relationships, as determined by the teams themselves. The teams studied in Project Aristotle ranged from three to fifty individuals (with a median of nine members).

Defining “effectiveness”

Once they understood what constituted a team at Google, the researchers had to determine how to quantitatively measure effectiveness. They looked at lines of code written, bugs fixed, customer satisfaction, and more. But Google’s leaders, who had initially pushed for objective effectiveness measures, realized that every suggested measure could be inherently flawed – more lines of code aren’t necessarily a good thing and more bugs fixed means more bugs were initially created.

Instead, the team decided to use a combination of qualitative assessments and quantitative measures. For qualitative assessments, the researchers captured input from three different perspectives – executives, team leads, and team members. While they all were asked to rate teams on similar scales, when asked to explain their ratings, their answers showed that each was focused on different aspects when assessing team effectiveness.

Executives were most concerned with results (e.g., sales numbers or product launches), but team members said that team culture was the most important measure of team effectiveness. Fittingly, the team lead’s concept of effectiveness spanned both the big picture and the individuals’ concerns saying that ownership, vision, and goals were the most important measures.

So the researchers measured team effectiveness in four different ways:

  1. Executive evaluation of the team
  2. Team leader evaluation of the team
  3. Team member evaluation of the team
  4. Sales performance against quarterly quota

The qualitative evaluations helped capture a nuanced look at results and culture, but had inherent subjectivity. On the other hand, the quantitative metrics provided concrete team measures, but lacked situational considerations. These four measures in combination, however, allowed researchers to home in on the comprehensive definition of team effectiveness.

Collecting data and measuring effectiveness

Using input from executives across the globe, the research team identified 180 teams to study (115 project teams in engineering and 65 pods in sales) which included a mix of high- and low-performing teams. The study tested how both team composition (e.g., personality traits, sales skills, demographics on the team) and team dynamics (e.g., what it was like to work with teammates) impact team effectiveness. Ideas were pulled from existing research as well as Google’s own experience with what makes an effective team.

They conducted hundreds of double-blind interviews with leaders to get a sense of what they thought drove team effectiveness. The researchers then looked at existing survey data, including over 250 items from the annual employee engagement survey and gDNA, Google’s longitudinal study on work and life, to see what variables might be related to effectiveness.

Here are some sample items used in the study that participants were asked to agree or disagree with:

  • Group dynamics: I feel safe expressing divergent opinions to the team.
  • Skill sets: I am good at navigating roadblocks and barriers.
  • Personality traits: I see myself as someone who is a reliable worker (informed by the Big Five personality assessment).
  • Emotional intelligence: I am not interested in other people’s problems (informed by the Toronto Empathy Questionnaire).

Demographic variables like tenure, level, and location were also collected.

Identifying dynamics of effective teams

With all of this data, the team ran statistical models to understand which of the many inputs collected actually impacted team effectiveness. Using over 35 different statistical models on hundreds of variables, they sought to identify factors that:

  • impacted multiple outcome metrics, both qualitative and quantitative
  • surfaced for different kinds of teams across the organization
  • showed consistent, robust statistical significance

Google's Team Dynamics

The researchers found that what really mattered was less about who is on the team, and more about how the team worked together.

In order of importance:

  1. Psychological safety: Psychological safety refers to an individual’s perception of the consequences of taking an interpersonal risk or a belief that a team is safe for risk taking in the face of being seen as ignorant, incompetent, negative, or disruptive. In a team with high psychological safety, teammates feel safe to take risks around their team members. They feel confident that no one on the team will embarrass or punish anyone else for admitting a mistake, asking a question, or offering a new idea.

  2. Dependability: On dependable teams, members reliably complete quality work on time (vs the opposite – shirking responsibilities).

  3. Structure and clarity: An individual’s understanding of job expectations, the process for fulfilling these expectations, and the consequences of one’s performance are important for team effectiveness. Goals can be set at the individual or group level, and must be specific, challenging, and attainable. Google often uses Objectives and Key Results (OKRs) to help set and communicate short and long term goals.

  4. Meaning: Finding a sense of purpose in either the work itself or the output is important for team effectiveness. The meaning of work is personal and can vary: financial security, supporting family, helping the team succeed, or self-expression for each individual, for example.

  5. Impact: The results of one’s work, the subjective judgement that your work is making a difference, is important for teams. Seeing that one’s work is contributing to the organization’s goals can help reveal impact.

The researchers also discovered which variables were not significantly connected with team effectiveness at Google:

  • Colocation of teammates (sitting together in the same office)
  • Consensus-driven decision making
  • Extroversion of team members
  • Individual performance of team members
  • Workload size
  • Seniority
  • Team size
  • Tenure

It’s important to note though that while these variables did not significantly impact team effectiveness measurements at Google, that doesn’t mean they’re not important elsewhere.

For example, while team size didn’t pop in the Google analysis, there is a lot of research showing the importance of it. Many researchers have identified smaller teams – containing less than 10 members – to be more beneficial for team success than larger teams (Katzenbach & Smith, 1993Moreland, Levine, & Wingert, 1996.

Smaller teams also experience better work-life quality (Campion et al., 1993), work outcomes (Aube et al., 2011), less conflict, stronger communication, more cohesion (Moreland & Levine, 1992Mathieu et al., 2008), and more organizational citizenship behaviors (Pearce and Herbik, 2004).

Helping teams determine their own needs

Beyond just communicating the study results, the Google research team wanted to empower Googlers to understand the dynamics of their own teams and offer tips for improving. So they created a survey for teams to take and discuss amongst themselves.

Survey items focused on the five effectiveness pillars and questions included:

  1. Psychological safety – “If I make a mistake on our team, it is not held against me.”
  2. Dependability – “When my teammates say they’ll do something, they follow through with it.”
  3. Structure and Clarity – “Our team has an effective decision-making process.”
  4. Meaning – “The work I do for our team is meaningful to me.”
  5. Impact – “I understand how our team’s work contributes to the organization’s goals.”

After completing the survey, team leads received aggregated and anonymized scores to share with team members and inform a discussion. A People Operations facilitator would often join the discussion, or the team lead would follow a discussion guide created by the People Operations team.

Read about the researchers behind the work in The New York Times: What Google Learned From Its Quest to Build the Perfect Team

Related Articles