The One-Paella Rule
- 2-minute read
Amazon has a famous pizza rule:
No team should be big enough that it would take more than two pizzas to feed them.
Numerous studies have shown the advantages of smaller teams, many predating Amazon. Members of a small team put in more effort because their contribution feels indispensable1 and the fruits of their labour are more visible. While a team requires sufficient expertise to solve complex problems, those that grow too large struggle to reach a consensus2. As the number of "links" between members grows quadratically with team size, individuals spend much more time coordinating and assessing how their role fits into a larger team. And in my experience, psychological safety – the bedrock for any sustainable collaboration – is more easily established in smaller groups.
In 2012, three professors performed an experiment where teams were tasked with recreating a Lego build3. Some teams had two members, others had four. Within each team, only one member was allowed to check the reference build at a time, necessitating coordination. Contrary to Jacob Morgan's misrepresentation of the data4, the four-member teams on average completed their build 28% faster than the two-member teams. However, that is still a 45% increase in person-minutes (labour). Interestingly, estimators correctly predicted the larger teams would be less efficient, suggesting we recognise the benefits of a small team (although they underestimated the increase at only 31%).
That said, as a true European with navy blue blood flowing through my veins and gold stars in my heart, I have an issue with the two-pizza rule. I am not one to forbid pineapple on pizza, and I respect that everything is bigger in Texas, but I insist that a classic pizza is plate-sized and serves a single person.
A two-pizza rule with traditional pizzas implies a two-person team. While duos might be optimal for building Lego sets, tech teams tend to have a scope too broad for two people. So let's reject the two-pizza rule, and embrace the one-paella rule: one team, one pan. Ζήτω η Ευρώπη! Aontaithe san éagsúlacht!
- Kerr, N. L., & Bruun, S. E. (1983). Dispensability of member effort and group motivation losses: Free-rider effects. Journal of Personality and Social Psychology, 44, 78–94.
- Hare, A. P. (1952). A study of interaction and consensus in different sized groups. American Sociological Review, 17, 261–267.
- Staats, B. R., Milkman, K. L., & Fox, C. R. (2012). The team scaling fallacy: Underestimating the declining efficiency of larger teams. Organizational Behavior and Human Decision Processes, 118, 132–142.
- In a 2015 article titled "Why Smaller Teams Are Better Than Larger Ones", Jacob Morgan wrote that "[…] the team comprised of 2 people accomplished the task in 36 minutes whereas the team comprised of 4 people finished the task in 56 minutes", which can be interpreted as the two-member teams being faster in clock time; when in reality they were merely more efficient ‒ averaging fewer person-minutes (labour).