What are the best analyses of small, innovative, productive groups?

Shane emails me:


What have you found to be the best books on small, innovative, productive groups?

These could be in-depth looks at specific groups – such as The Idea Factory, about Bell Labs – or they could be larger studies of institutions, guilds, etc.

I suggest reading about musical groups and sports teams and revolutions in the visual arts, as I have mentioned before, taking care you are familiar with and indeed care passionately about the underlying area in question.  Navy Seals are another possible option for a topic area.  In sociology there is network theory, but…I don’t know.  In any case, the key is to pick an area you care about, and read in clusters, rather than hoping to find “the very best book.”  The very theory of small groups predicts this is how you should read about small groups!

But if you must start somewhere, Randall Collins’s The Sociology of Philosophies is probably the most intensive and detailed place to start, too much for some in fact and arguably the book strains too hard at its target.

I have a few observations on what I call “small group theory”:

1. If you are seeking to understand a person you meet, or might be hiring, ask what was the dominant small group that shaped the thinking and ideas of that person, typically (but not always) at a young age.  Step #1 is often “what kind of regional thinker is he/she?” and step #2 is this.

2. If you are seeking to foment change, take care to bring together people who have a relatively good chance of forming a small group together.  Perhaps small groups of this kind are the fundamental units of social change, noting that often the small groups will be found within larger organizations.  The returns to “person A meeting person B” arguably are underrated, and perhaps more philanthropy should be aimed toward this end.

3. Small groups (potentially) have the speed and power to learn from members and to iterate quickly and improve their ideas and base all of those processes upon trust.  These groups also have low overhead and low communications overhead.  Small groups also insulate their members sufficiently from a possibly stifling mainstream consensus, while the multiplicity of group members simultaneously boosts the chances of drawing in potential ideas and corrections from the broader social milieu.

4. The bizarre and the offensive have a chance to flourish in small groups.  In a sense, the logic behind an “in joke” resembles the logic behind social change through small groups.  The “in joke” creates something new, and the small group can create something additionally new and in a broader and socially more significant context, but based on the same logic as what is standing behind the in joke.

5. How large is a small group anyway?  (How many people can “get” an inside joke?)  Has the internet made “small groups” larger?  Or possibly smaller?  (If there are more common memes shared by a few thousand people, perhaps the small group needs to be organized around something truly exclusive and thus somewhat narrower than in times past?)

6. Can a spousal or spouse-like couple be such a small group?  A family (Bach, Euler)?

7. What are the negative social externalities of such small groups, compared to alternative ways of generating and evaluating ideas?  And how often in life should you attempt to switch your small groups?

8. What else should we be asking about small groups and the small groups theory of social change?

9. What does your small group have to say about this?

I thank an anonymous correspondent — who adheres to the small group theory — for contributions to this post.


'ask what was the dominant small group that shaped the thinking and ideas of that person'

'Are you now, or have you ever been, a member of a chess club?' sounds like a potentially very revealing question.

You might reveal whether they like chess or not. That's about it.

Or, with that phrasing, it might revial more about the person asking that question than the person answering it.

Interviews are actually two way streets, assuming the person being interviewed is not in desperate circumstances.

What were the chances Kurt Cobain and Dave Grohl would be in the same band?

I would guess that small groups select for those that stand out in the enthusiast milieu of various industries. Consider that Steve Jobs and Woz met through homebrew enthusiasts groups. It’s dificult for large corporations to replicate that.

"What were the chances Kurt Cobain and Dave Grohl would be in the same band?"

Pretty high. It's a little like what were the chances of Ringo Starr being in the same band as Lennon and McCartney? Ringo was the best professional drummer in Liverpool when the Beatles were ready to make it big.

Grohl joined Nirvana fairly late, at a point when a canny music industry individual like Grohl could see that Cobain was just a kick-ass drummer short of making it big.

(By the way, my impression is that Dave Grohl is the opposite personality of Cobain: Grohl is a high highly effective, positive individual who is very good at making things better for himself and those around him.)

Look at the best things that came out of the USA in the 1950's & 1960's

e.g., Kelly Johnson's 'Skunk Works' @ Lockheed

& Chris Kraft's Mission Control [his own book OOP, but Gene Kranz's is available]:


"Skunk Works," the memoir of Ben Rich, Kelly Johnson's successor at Lockheed during the stealth era, is a wonderful book.

Tyler is just recreating the conflict between hierarchy and markets or order and variety that Coase focused on in his original paper on the firm. Williamson and others have elaborated on this within the firm. A firm that is hierarchical and dictatorial has order and decisiveness but loses some variety and competition. On the other hand, too many competing small groups with too much freedom can lead to too much rivalry and in the end destroy the existence of the overall group or organization itself. Which system is efficient is dependent on the size of the overall group, the nature of the desired product/outcome including related technology, and the nature of the external environment of competition/market/technology faced by the firm. North took this logic and developed it for the State. And North, Wallis, and Weingast as well as work, by Olson, Nye, and others struggled to reconcile the logic or organization while dealing with the problems of force, coercion, and necessary order.

I should add that there is a literature on communal/worker-owned firms and why they only flourish under very stringent conditions and usually below a certain size.

Whenever Bill Gore had a hit product that required increasing his headcount above 160 people to meet demand he always did one thing:

Build another plant.


As Cowen suggests, the dominance of media (Cowen mentions the internet but media extends well beyond the internet) in our lives has made small groups larger, or stated another way, made the small group potentially much more influential, including small groups of offensive people or stupid people. Moreover, the dominance of media has emboldened small groups to insist on public platforms to express their views, no matter how offensive or stupid, equating the right of free speech to the right to the public platform. [By "public platform" I am not referring only to the street corner but to forums that provide for a wide distribution of what is being expressed including public places such as the university and media of all types.] Does Cowen agree that the small group has a right to the public platform, or does Cowen believe the small group only has the right to express their views? I assume from his many comments that it's the former, that the small group has the right to the public platform to express their views no matter how offensive or stupid. That would be the Millian (John Stuart Mill) view of absolute free speech. But in Mill's day, media was far from being dominant in peoples' lives, and the public platform (in the broad sense) for expressing offensive or stupid views was neither a right nor even available. By equating the absolute right to free speech with the right to express it in a public platform, are we making the public both more offensive and moronic? After all, views, no matter how offensive or stupid, when broadly expressed are much more likely to be accepted as correct by far more people by virtue of the public platform.

Skunk Works

My guess is that small groups with big backing can do a lot....

I hate to say it, but the need to survive (free markets, destruction) might suffocate a lot of small groups in the crib....

I'm not sure that the Lockheed 'Skunk Works', the Xerox PARC (not previously mentioned), or Bell Labs fit easily into this. All three of these 'small groups' were backed by funding from large corporate enterprises. Bell Labs had the distinct advantage of being backed by a monopolistic parent and look at what happened when ATT was broken up. Bell Labs was part of Lucent for a while until the company decided to end non-applied research and Bell Labs has been part of Nokia since 2016.

Another model of corporate funded small group research was the Roche Institute of Molecular Biology that was established in the late 1960s when Roche was making scads of money from Valium and Librium. The institute was modeled more like an academic campus with many scientists recruited straight out of academe and NIH.

HBR has a case study on the Whitesides (chemistry) lab.

8. What is the common objective of this group? Is there more than one? Which relates to 5.

Tune In, Mark Lewisohn

The best guideline is Jeff Bezos' "Two Pizza Rule": "Never have a meeting where two pizzas couldn't feed the entire group...The idea is that most attendees will end up agreeing with each other instead of voicing their own opinions and ideas."

Restaurants and chefs: The Art of the Restaurateur by Nicholas Lander, Life on the Line by Grant Achatz and Nick Kokonas, Beer School by Steve Hindy and Tom Potter

Startup Communities by Brad Feld

Beethoven for a Later Age by Edward Dunsinberre

The Driver by Alex Roy

I think 6 has potential; the question is how to aggregate the pairs.

How about Trump Casino and Resorts? Oh wait, the multiple bankrutpcy thing. How about Trump University? Oh wait, the defrauding students thing.

As a software engineering manager, I spend a lot of time thinking about how to make small teams innovative and productive. Some sources I've found useful include:

Social Physics by Alex Pentland. Prof Pentland of the MIT Media Lab used Big Data to study work groups. He found that the most important determinant of group success was evenness of communications. That is if each member of the group talked about the same amount that groups was most successful. Groups with this characteristic outperformed groups containing a superstar and outperformed groups where the average ability was much higher.

To understand why small groups might outdo big groups, it's worth reading the essay The Mythical Man Month by Fred Brooks. This is one of the most famous essays in software engineering written by the leader of the IBM team that developed the first modern operating system. It points out that each person added to a team increases the teams communications overhead. So anything a extra person brings to a team has to be weighted against the communication cost.

Lean, the organizational system developed by Toyota that is now used in everything from construction to hospitals to software engineering, is based entirely on small teams. The Toyota Way by Liker is a good introduction to this method. To see how it works for startups, try Lean Startup by Eric Ries.

Ideo is a widely admired industrial design firm that developed a method called Design Thinking that its small groups use to innovate. These methods are now widely used by software development teams. This is a good introduction: https://dschool.stanford.edu/resources-collections/a-virtual-crash-course-in-design-thinking

Finally this New York Times article summarizes what Google learned when it studied what made for successful teams: https://www.nytimes.com/2016/02/28/magazine/what-google-learned-from-its-quest-to-build-the-perfect-team.html

Comments for this post are closed