When you’ve got a small group, you don’t need to constantly formalize things. You communicate and you know what’s going on. If you have a question about something, you ask someone. Formalized rules, deadlines, and documents start to seem silly. Everyone’s already on the same page anyway.
Ten-groups
According to British author Antony Jay, there are centuries of evidence to support the idea that small groups are the most efficient. In “The Corporation Man,” he talks about how humans have worked in small groups, usually five to fifteen people, as hunters and farmers for hundreds of generations. The ideal group size is a ten-group:
He found the most efficient to be organised in groups of eight to fourteen people which he came to call ‘ten-groups’, each group free to find its own way towards a target set for it within the general objects of the corporation…
“The basic unit is [a group] which varies from three to twelve or fifteen in number, and perhaps optimizes somewhere around ten; that this group is bound together by a common objective, and that the bond of trust and loyalty thus formed can become an extremely powerful uniting force; that the group needs to decide on (or at least take part in deciding on) its own objective, and to work out for itself how that objective shall be achieved…”
He offers up interesting examples to back up the theory, from sports teams to juries to army squads:
Jay draws attention to units of around this size in many fields beyond the corporation. A committee works best with about ten members; if it grows much beyond that size the extra people do not take a fully active part. Nearly all team games use a group of about ten on each side. Juries have 12 members and the Jewish minyan 10. In an army, organization often decides life and death, and under this pressure armies, too, adopt a basic unit of about ten; the British army, the US army, the ancient Roman army and that of Genghiz Khan, in fact every long-standing successful army, has built up its larger formations from squads or sections of about this size.
That mention of the Roman army takes us back some two thousand years, and Jay traces the ten-group back still farther, back to the foraging communities. The ten-group, found today as a structural unit in successful corporations began, he argues, as the male hunting-group of pre-agricultural times, still with us and still functional.
Groups this size succeed because they have mutual dependence and a common objective:
This group displays qualities in addition to its size. Small enough for the contribution of each member to make a noticeable contribution, in order to function it needs mutual dependence, a common objective and a single criterion of success for them all; as the hunting band fed or went hungry together so members of the modern ten-group must receive praise, blame and material rewards collectively for the unit to function at its best.
Two-pizza teams
The tech world equivalent of ten-groups: two-pizza teams. Here’s what Marc Hedlund says about two-pizza teams:
I believe both Apple and Amazon (two companies I respect a lot for their version 1.0s) talk about “two-pizza teams” — that is, keeping the team small enough that it can be fed with two pizzas…If the company that made the iPod believes in a two-pizza team, how can you justify putting 90 people in the room?
Bottom line
If you’re working in a group bigger than 15 people, you’re fighting human nature. Start a satellite group and see if you can lose the busywork and bureaucracy that inevitably accompany oversized teams.
Related
Some other Antony Jay quotes:
He’s suffering from Politicians’ Logic. Something must be done, this is something, therefore we must do it.
‘Referring the matter to a committee’ can be a device for diluting authority, diffusing responsibility and delaying decisions.
The uncreative mind can spot wrong answers, but it takes a very creative mind to spot wrong questions.
You can judge a leader by the size of the problem he tackles. Other people can cope with the waves, it’s his job to watch the tide.
You can stop focusing on extraneous collateral and pay attention to what matters instead.
Peter Urban
on 24 Apr 08I agree that small groups are way better in terms of getting (smallish) things done. What if you have to do something huge, like build an hydrogen airliner? I’d love to learn how large organizations can become more efficient and fun to work at) by clustering their projects into smaller groups and applying the ‘getting real’ rules and still be able to pull something big off as a collective. Any insights form boeing?
David Barrett
on 24 Apr 08What size pizzas are these? Over here, two pizza teams are two people :)
Eoghan McCabe
on 24 Apr 08My thought exactly, Dave! In fact, I knew a guy once that was a two-man team all by himself. :-)
Erik Peterson
on 24 Apr 08I wrote a blog post about this not too long ago, and came up with five as the ideal size.
My theory on five is that with a group of that size, there is still room for discrete and tightly defined roles amongst the members. Anything larger than that (and, certainly, at groups as large as 10), and individuals lose role identity. I believe it is the role identity that keeps people productive- in large groups that identity wanes along with productivity.
tyler rooney
on 24 Apr 08Malcolm gladwell have a great piece about keeping teams/offices small in The Tipping Point when he talks about Gore (as in GoreTex, not Al Gore) and how they keep all of there offices smaller than 150 people. Here’s the excerpt from Google Books.
David, they’re two very large pizzas.
Daniel Tenner
on 24 Apr 08Totally agree. In fact, I think 10 is a bit too big for software development teams.
I personally believe the ideal size for a software development team is between 3 and 5 developers. I guess this doesn’t leave any room for all the other things that need to happen to launch a project, but if you have, say, 3 developers and a product manager, if you can develop your project with that, then you shouldn’t take on any more people unless you really have to.
Currently my project has 2 developers and 1 product guy. That is a little on the small side, but with such a small team we are, I believe, ruthlessly effective, even though we’re spread across 2 continents and don’t have an office!
Daniel
Sean
on 24 Apr 08Paul Graham wrote an article “You Weren’t Meant to Have a Boss” that makes similar points.
condor
on 24 Apr 0812 disciples. case closed.
jan korbel
on 24 Apr 08Speaking of team work, here’s a piece in NY Times nodding to your long standing point about co-working from distance. http://tinyurl.com/4o8hk7
Thomas
on 24 Apr 08“If the company that made the iPod believes in a two-pizza team, how can you justify putting 90 people in the room?”
Well, cause they didn’t use a two-pizza team to design or make the iPod.
engtech
on 24 Apr 08I’m a two pizza team all by myself :(
John D. Mitchell
on 24 Apr 08The two pizza “rule” really depends on where you get your pizzas and the appetite of your team members. :-)
FWIW, a “one pizza” group of e.g., 4-5 people is what the armed forces have called “fire teams”. A two pizza team is a squad.
Groups of up to about 150 people are called companies in the military but a better way to think about them, IMHO, is that that’s what a village is. That’s basically the limit as to the number of people you can have a relatively personal, direct relationship with.
Laurel
on 24 Apr 08Interesting that when I think of a group of 10, I think “wow, that’s a huge group”.
I think the kind of decision making process used affects the group size. I think the optimal group size is larger when there’s a strong leadership role and possibly a hierarchical structure (large companies, military, stereotypical cave man hunters), but when something closer to consensus decision making is used the optimal group size might be a lot closer to 4 (group of friends traveling, some startups, family, nonprofit group).
It’s interesting that the linked article specifically talks about male groups. Maybe if the objective and organization of the group is “male” (a lot of the examples seem to be about killing stuff) the group size is different than if it’s “female”.
Anthony Morales
on 24 Apr 08I’m w/ Peter Urban. Same questions. I’m in a large company and have tried to keep my groups small (3-10). The problems arise when sign-off has to come from multiple groups or stakeholders. Their peers or reports start voicing an opinion (helpful or otherwise). In large projects (like the one I’m in now), we get bogged down. So, for lack of a better question: how does this idea scale?
Ben Lacy
on 24 Apr 08Fantastic post. I’ve always believed in this theory. There need to be more companies like 37signals that have small teams and cater to a decent-sized target audience. We don’t need anymore massive companies that try to create a single product to answer everybody’s needs. That just clutters things up.
Why not have several choices that are more tailored to specific uses? Don’t force the masses to choose between a couple ridiculously huge and complex solutions that aren’t any fun to use at all.
Cheers.
Randy Henne
on 24 Apr 08I used to run a 2 pizza team (or 2PT) at Amazon. We joke a lot about the pizza and how many we could eat too.
The main ideas about the 2PT were that they were relativley self-sufficent and independant. The 2PT idea seemed to fit best with nicely containable areas like strictly innovation areas. My area for instance was a specific piece of the Amazon personalization, data mining and recommendations puzzle.
Another key component is that the teams are measured by well defined metrics but allowed to choose there own path to get there.
I agree with some earlier posts, this works well with innovation projects but a very large coordinated effort may not fit that well.
Ro
on 25 Apr 08On-point post Matt.
Peter and Anthony, what if you have to invade and occupy an entire country for five years? If you’re the Roman, US, or practically any other military, you use10-groups “army squads”—so the article kind of directly answers your question.
If you’re looking for biz world examples, there is one in Gore-Tex’s parent company, which has grown to 3,000+ employees and takes on large scale manufacturing and marketing projects…. in small groups.
John S. Kim
on 25 Apr 08good read Matt.
Although I agree to the point, I still have to wonder, would it be possible to have big game development teams to be sized down to 10? or rather, how can large game development teams be organized so that people will still be effective and efficient?
I don’t think games like World of Warcraft can be made with 10 people. Curious.
Jeff Sullivan
on 25 Apr 08The big companies building big projects that divide functionality among smaller groups, suffer more from Conway’s Law than a single small group. The garbage produced from such large organizations can be truly horrendous.
Wesley Anderson
on 25 Apr 08I think the “Two Pizza Team” thing means up to 16 people. Enough for one slice of pizza per person in a brainstorming session.
Llewelyn Roderick
on 25 Apr 08Great post and if you want another angle on this idea check out Malcolm Gladwell’s “The Tipping Point”. In looking at how information spreads in an organisation or group he identifies the number 150 (Dunbar’s number) and cites the Goretex company as proof of concept. Goretex don’t have any single business unit larger than 150 people. As soon as a division grows beyond that number they build a new building and hive it off! The net effect is that Goretex has no corporate hierarchies. Instead the groups self-regulate as they see fit.
Chance Bliss
on 25 Apr 08I love this post. I’m in the middle of a culture clash between 2 recently merged companies. The one I’m in has roughly 5 team members per application. We maintain a low level of documentation (mainly help, faqs and network/hardware specifications) and process using Basecamp and a Wiki.
The other organization is comprised of 150+ team members (outsourced to India) maintaining just one application. They are heavy on documentation, process and approvals. Since the merger 6 months ago, we have completed four major releases. They have completed zero.
This is not to bash the other team. They are all hard working folks, but a huge amount of time is wasted just managing the communications between all those developers. While I can just turn my head left or right and get an immediate answer. No conference calls and WebExs necessary.
RedesignYourBiz.com
on 25 Apr 08nice article…... thanks
Jeff Sullivan
on 25 Apr 08“They are all hard working folks, but a huge amount of time is wasted just managing the communications between all those developers.”
Like I said, it’s called Conway’s Law
- Organizations which design systems are constrained to produce designs which are copies of the communication structures of these organizations.
Or more concisely:
- Any piece of software reflects the organizational structure that produced it.
Tim
on 25 Apr 08So, 10-12 people would be some kind of maximum size. What about minimum size? Do you think there needs to be at least 5, 7, less? people to be the most effective?
I feel that having a group to start with helps pulling each other forward, thus pulling the whole group…
I liked the quote “two companies I respect a lot for their version 1.0s”. I never thought about successful companies this way, and I find that an interesting way of looking at quality/success of companies (maybe also people too). It can’t define their current value but it does indeed show some positive characteristics and values.
rsomers
on 27 Apr 08I find that ‘fire teams’ (John D. Mitchell’s word) of 4-5 are most effective in cutting through organizational red tape because everyone involved feels committed. By the time it hits 10 or more, there are always a couple of free riders.
Anthony Morales (hi, Ant!) has a good point above in that it’s often not the size of the project team that’s the problem. The size of the population of real (or, more often, perceived) stakeholders is the issue.
rsomers
on 27 Apr 08One other illustration: even the Borg know this principle. She was Seven of Nine, not Seventy-Three of Nine Hundred and Eighteen.
I am SUCH a geek…
herve
on 28 Apr 08the biggest problem in large corporation is for 1 person who has the power to say yes, you have 15 who has only the power to say no. Remember : never underestimate the power of stupid people in large group So yes, If you’re working in a big group and you want to make things happen, you’re fighting human nature
Calvin Bottoms
on 28 Apr 08And then there were Jesus and the twelve.
Nathan Zook
on 28 Apr 08On the issue of scaling: this principle certainly applies as one goes up the corporate hierarchy, except the teams get smaller. One of the better Napoleonic games I played had a limit on the number of “armies” a general could command before his effectiveness started to decline. The genius Napoleon could control six.
Military organizations tend to be pretty narrow—three X in a Y, but this has a lot to do with the fact that in many areas, an army is a body shop, and you need flexibility in the size of teams that you deploy. Also, attrition can be an issue.
More generally, a small group of team leaders can agree on how to break up a larger problem, and return to their teams—where the smaller problem is now broken up further.
Finally, do note that there is evidence of significant internal friction in the twelve of Jesus. In fact, he had an inner circle of three.
This discussion is closed.