Dunbar’s Number, Psychological Safety and Team Size

Dunbar’s Number

Robin Dunbar, British anthropologist and evolutionary psychologist, in 1993 studied brain size, group size, and language development in humans and primates, and developed the concept of “Dunbar’s Number” – 150. This, in theory, is the approximate maximum number of people that we humans can concurrently maintain stable relationships with (I’m assuming most newsletter readers are human, but we welcome all species!). Dunbar suggested that a maximum group size of 150, which he found represented in natural-forming groups all over the world in different domains and cultures – from tribes to military units – is a function of neocortex size in the primate brain. Dunbar’s book “Human Evolution” is an excellent write up of the social and cognitive evolution of our species if you’re interested further.
 

Dunbar also described these smaller units of the “support clique”, “sympathy group”, and “active network” as well as the personal network limit of 150 (the red dot in the middle of the image above is the individual or ego state). He was studying the total number of relational connections that we can maintain, but his theories have been applied (as far as I can tell) largely to teams and organisations, and Facebook friends. I’m not sure where this leaves our connections with friends and family outside our work groups… 🙂

Dunbar’s Number in Practice

As Malcolm Gladwell describes in The Tipping Point, companies such as Gore (the firm behind Gore-Tex, to whom every British hiker owes a debt of gratitude!) famously capped their factories at 150 people, because beyond that number, Bill Gore observed that people began to lose a sense of community – they could no longer remember faces and names of all the people in the building. In the military, Dunbar’s boundaries do appear to manifest: teams (2-4 members), squads (6-12), platoons (around 50), companies (100-200), etc, but there’s a huge variance and it’s not clear that these boundaries remain consistent.

As much as Dunbar’s limits on group sizes might seem to be common sense, and reflected in many real world examples, Dunbar’s theories on group size boundaries have been deconstructed and shown to possess confidence intervals too large to be robust in the real world. That is, group size boundaries do exist, but may be anywhere from 30 to 250, depending on context, culture, and other factors. “Dunbar’s assumption that the evolution of human brain physiology corresponds with a limit in our capacity to maintain relationships ignores the cultural mechanisms, practices, and social structures that humans develop to counter potential deficiencies.” Ruiter et al, 2011.

All Models are Wrong and Some are Useful

For me, Dunbar’s Number falls into the “wrong but useful” set of theories and models that we can use to work with team size. It’s certainly fair to say that we can maintain closer relationships with a small group of people, whilst we have extending “circles” of friends, acquaintances and colleagues. 

Communication Complexity

Part of this is related to communication complexity, which increases exponentially with group size, as you can see in the diagram below. As group size increases, it becomes increasingly difficult to maintain communications with everyone else in the group. And contrary to how it may instinctively seem, adding one more person to an already large team has a greater impact on communication that adding someone to a smaller team. Indeed, as we’ve previously mentioned in the newsletter, Brooke’s Law states that “Adding people to a late project makes it even later.“

Of course, we can mitigate this effect with effective communication tools, processes, team charters and more, but team size is, and will always be, a factor in team performance and cohesion.

This diagram does a great job of showing how as group size increases, the number of 1-1 relationships, or lines of communication, grows exponentially. This can be described via a formula, where r is the number of potential 1:1 relationships and n is group size:

r = (n x (n-1)) / 2

5 people gives the following result;

(5 [group size] x 4 [group size minus one]) / 2 = 10 [potential relationships]

10 people gives the following result;

(10 [group size] x 9 [group size minus one]) / 2 = 45 [potential relationships]

50 people gives the following result;

(50 [group size] x 49 [group size minus one]) / 2 = 1225 [potential relationships]

1225 potential relationships! Doubling from a group of 5 to a group of 10 increases relationships by x4.5, and going up to 50 increases by x27.2!

This is closely related to Metcalfe’s Law in telecommunications – the value of a network goes up as a function of the square of the number of users.

Jeff Bezos famously adhered to Amazon’s “two-pizza-team” rule as a way to limit team sizes and maintain the effectiveness of small, tight-knit teams: No team should be larger than the number of people that can be adequately fed by two large pizzas. However, whilst this is effective for teams that can truly own a value stream, it appears less crucial for functional teams that have complex and unavoidable organisational dependencies. It seems that biggest predictor of a team’s success wasn’t whether it was small, but whether it had a leader with “the appropriate skills, authority, and experience to staff and manage a team whose sole focus was to get the job done.” 

Psychological Safety and Team Size

Very little research has been carried out to directly examine the effect between team size and psychological safety, however, as we’ve seen above, there is a great deal of evidence that suggests that optimum team sizes do exist, within certain parameters and dependent upon other variables. Studies on psychological safety often have to control for team size in order to detect mediating effects of psychological safety on team effectiveness and job satisfaction, for example. In other studies on management teams, researchers found a negative association between team size and team performance and that team psychological safety significantly mediated this relationship.

As Amy Edmondson points out in her 2004 paper, “the increased number of relationships and complex interdependencies in a very large group may diminish the amount of repeated interaction between any subset of members and thereby reduce the degree to which consistent perceptions of psychological safety take shape.”

That is to say, as a member of a group, predicting the “risk” of interpersonal consequences of speaking up is an easier mental calculation in a group of 3 versus a group of 30, or even 300. Simply, psychological safety is related strongly to the predictability we feel about the response to anything we do or say in a group. That predictability decreases as group size increases partly because it’s simply harder to get to know everyone well enough, and partly because the chances of someone being a “bad actor” in that group increases with group size.

Essentially: The chances of someone punishing or humiliating us as a result of speaking up increases as group size increases: as a result, we tend to feel psychologically safer in smaller groups. 

SLLLAM Teams

The best performing teams are generally Small, Lean, Long-Lived, Autonomous, and Multi-disciplinary – and psychologically safe. It’s fair to say that psychological safety is generally more difficult to foster as group size increases.

So, we can use theories such as Dunbar’s numbers, communication complexity, and 2-pizza rules to help inform decisions around team size – but as usual, one-size-fits-all fits everyone poorly: there is no “perfect” team size.

This newsletter is sponsored by Candor.

Candor is the platform to design and exchange beautiful personal readmes with your team. You’ll be able to express how you work and help your team see how to get the best from you. It’s meaningful, fun, and unashamedly human. The best bit? Team psychological safety is fostered through team members better understanding each other and how to work together.  

Candor has helped teams build trust and empathy at Google, DoorDash, Carta, Lattice, and 500+ startups.

It’s 100% free to join. Sign up here. 
 

Psychological Safety In the Workplace:

Here’s a great podcast episode from Sal Silvester, with Amy Edmondson, who happen to be discussing topics that I’ve been working on with groups this week, including the relation of hierarchy to psychological safety, and actual physical safety. 

Steve Ingham, sports and performance scientist, does a great podcast called Supporting Champions, and was kind enough to invite me on recently to talk about psychological safety in high performance sports, or at least how my work relates to it.
 

Here’s a good piece from Kate Arms at Psychological Safety at Scale, about complaints being opportunities in disguise: “We often complain when we can’t articulate what’s bothering us.“

I love this from Sonia Sparkles – the psychological safety pizza! How many slices do you have?
 

Things to do and try:
Whilst this example is taken from primary (elementary) school teaching practice, I use it in my workshops with grown-ups and it’s great! The KWL gridallows participants to show what they already know about a topic (K, for Know), what they Want (W) to know, and then show what they have Learned (L). 

 The KWL grid helps participants to feel like they are coming to something with existing knowledge, and fosters a greater sense of confidence at the start of a session, as well as providing the facilitator or trainer with an idea of where to start from, where to go, and how successful the session has been (and where to go next). I hope you find it useful.

This week’s poem:

On The Ning Nang Nong, by Spike Milligan.

On the Ning Nang Nong
Where the Cows go Bong!
and the monkeys all say BOO!
There’s a Nong Nang Ning 
Where the trees go Ping!
And the tea pots jibber jabber joo.
On the Nong Ning Nang 
All the mice go Clang 
And you just can’t catch ’em when they do!
So its Ning Nang Nong
Cows go Bong!
Nong Nang Ning
Trees go ping
Nong Ning Nang
The mice go Clang
What a noisy place to belong
is the Ning Nang Ning Nang Nong!!

Listen to Spike Milligan himself read On The Ning Nang Nong in this video from 1995.

Mission statement

The mission of this newsletter is to make the world of work a better, safer, more inclusive and equitable place. We work towards this by sharing, as widely as possible, new research and opinion, case studies, news, practices and methods related to psychological safety. 

This includes topics such as inclusion, human factors, facilitation, safety critical environments, systems thinking, leadership and management, diversity and equity, mental health, organisational dynamics, culture, ethics, and more. 

A core goal is to amplify the voices of minoritised groups and those who are less represented in this field. This includes (but is not limited to) voices of people of colour, neurodiverse people, LGBTQ+ people, women in leadership and tech, and more.

Another goal is to share insights and best practices across different domains: for example, so that those of us in tech may learn from healthcare, and those of us in education can learn from aviation.

I believe knowledge is worthless unless it’s accessible. A key principle of this newsletter is to only share content that everyone can access. This excludes content behind paywalls, research papers that are not open-access, and proprietary tools that don’t open source their algorithms (preventing us from observing how systems make decisions about us). There may be some exceptions to this principle in practice, for example, where authors are able to share original texts of papers.

Whilst this statement will evolve this over time, the core principles of equity, amplification, and progressive & humanistic leadership shouldn’t change.