Write a Review

Let’s start with a few examples of very common real-world coordination problems.

  • The marketing department at a car dealership posts ads for specific cars, but the salespeople don’t know which cars were advertised, causing confusion when a customer calls in asking about a specific car. There’s no intentional information-hoarding, it’s just that the marketing and sales people don’t sit next to each other or talk very often. Even if the info were shared, it would need to be translated to a format usable by the salespeople.
  • Various hard problems in analysis of large-scale biological data likely have close analogues in econometrics. The econometricians have good methods to solve the problems, and would probably be quite happy to apply those methods to biological data, and the bio experimentalists would love some analytic help. But these people hardly ever talk to each other, and use different language for the same things anyway.
  • When the US invaded Grenada in the ‘80’s, the marines occupied one side of the island and the army occupied the other. Their radios were not compatible, so if an army officer needed to contact their counterpart in the marines, they had to walk to the nearest pay phone and get routed through Fort Bragg on commercial telephone lines.
  • Various US intelligence agencies had all of the pieces necessary to stop the 9/11 attacks. There were agencies which knew something was planned for that day, and knew who the actors were. There were agencies which knew the terrorists were getting on the planes. There were agencies which could have moved to stop them, but unfortunately the fax(!) from the agencies which knew what was happening wasn’t checked in time.
  • There are about 300 million people in the US. If I have a small company producing doilies, chances are there are plenty of people in the US alone who’d love my doilies and be happy to pay for them. But it’s hard to figure out exactly which people those are, and even once that’s done it’s hard to get them a message showing off my product. And even if all that works out, if the customers really want a slightly different pattern, it’s hard for them to communicate back to me what they want - even if I’d be happy to make it.

So coordination problems are a constraint to production of all kinds of economic value. How taut are those constraints?

Well, let’s look at the market price of relaxing coordination constraints. In other words: how much do people/companies get paid for solving coordination problems?

When I think of people whose main job is to solve coordination problems, here are some occupations which spring to mind:

  • Entrepreneurs’ main job is to coordinate salespeople, engineers, designers, marketers, investors, customers, regulators, suppliers, shippers, etc…
  • Managers’ main job is to coordinate between their bosses, underlings, and across departments
  • Investment bankers coordinate between investors, companies, lawyers, and a huge number of people within each of those organizations
  • Real estate developers coordinate between builders, landowners, regulators, renters, and investors

Note that all of these are occupations typically associated with very high pay. Even more to the point: within each of these occupations, people who solve more complicated coordination problems (e.g. between more people) tend to make more money. Even at the small end, the main difference between an employee and a freelancer is that the freelancer has to solve their own coordination problem (i.e. find people who want their services); freelancers make lots of money mainly when they are very good at solving this problem.

Similarly with companies. If we go down the list of tech unicorns, most (though not all) of them solve coordination problems as their primary business model:

  • Google matches company websites to potentially interested users
  • Facebook is a general-purpose coordination platform
  • Amazon and Ebay are general-purpose marketplaces: they match buyers to sellers
  • Uber/Lyft are more specialized marketplaces

Again: solving coordination problems at scale offers huge amounts of money.

This suggests that coordination problems are very taut constraints in today’s economy.

It’s not hard to imagine why coordination problems would be very taut today. Over the past ~50 years, global travel/transportation has gone from rare to ordinary, and global communication has become cheap and ubiquitous. Geographical constraints have largely been relaxed, and communication/information processing constraints have largely been relaxed. The number of people we could potentially coordinate with has expanded massively as a result - a small doily business can now sell to a national or even global customer base; a phone app can connect any willing driver in a city to any paying rider.

Yet human brains have not changed much, even as the number of people we interact with skyrockets past Dunbar’s number. It’s hard for humans to coordinate with thousands - let alone billions - of other humans. The coordination constraint remains, so as other constraints relax, it becomes more taut.

What Would This Model Predict?

One prediction: suppose I’ve decided to become a freelancer/consultant. I can invest effort in becoming better at my object-level craft, or I can invest effort in becoming better at solving my coordination problem - e.g. by exploring marketing channels or studying my target market. Which of these will make more money? Probably the latter - coordination constraints are very taut, so there’s lots of money to be made by relaxing them.

More generally, when evaluating new business ideas, questions on my short list include:

  • How will this business find people who would want to buy the product?
  • How will this business make those people aware of the product?

To the extent that coordination is an unusually taut constraint, answers to these questions will be the main determinant of business profitability - even more so than product quality.

Coming from a different direction, when considering a business idea we should ask how many different kinds of people this business needs to coordinate. At one point I worked at a mortgage startup, where the list of internal departments included marketing, sales, underwriting, legal, and capital markets on the mortgage side, plus design, engineering, and ops on the tech side, and on top of that we had to interface to at least a dozen external companies on a regular basis. Coordination is the primary constraint at a company like that.

Yet another direction: if coordination constraints are very taut, then we expect adoption of technology which makes coordination easier. One form of this is outlined in From Personal to Prison Gangs: people make themselves easier to coordinate with, by following standardized patterns of behavior and fitting into standard molds. For instance, in large organizations (where more people need to coordinate) we tend to see group-based identity: rather than understanding what John or Allan does, people understand what lawyers or developers do. Interactions between people become more standardized, and roles more rigid - these are solutions to coordination problems. Such solutions entail large tradeoffs, but coordination constraints are very taut, so large tradeoffs are accepted.

Conversely, if we want a world with less pressure to standardize behavior, then we need some other way to relax coordination constraints - some technology which helps people coordinate at scale without needing to standardize behavior as much. Such technology would probably see wide adoption, and potentially make quite a lot of money as well.


I often hear people they’d like object-level skill/effort to be rewarded more than marketing/sales, or they’d like to see less pressure to standardize behavior, or they’d like the world to be more individualized and identity to be less group-based. To the extent that we buy the picture here, all of these phenomena are solutions to coordination problems. Society rewards marketing over object-level skill, and tries to standardize behavior, because coordination constraints are extremely taut.

If we want the world to look less like that, then we need alternative scalable technologies to solve coordination problems.


12 comments, sorted by Highlighting new comments since Today at 2:02 PM
New Comment

Anecdote: Mealsquares has to interface with around 17 companies.

Love your posts! I’d recommend publishing them at a slower pace so people get a chance to see and read them. It’s also a bit confusing on LW when several posts are published on the same day, and you can’t tell which one is first in the series (just from the front page).

I also liked these posts but felt fairly overwhelmed by them all showing up at once.

An additional point worth noting is that there is tremendous social value in reducing coordination costs, but it's nearly impossible to capture that value, so it's very under-provided.

What does lowering coordination costs look like? Trade meetups, conferences, and similar events or locations to foster communication and coordination (Like EA and LW meetups and forums,)  as well as trustworthy information sharing - which is costly to individuals and mostly benefits others. (Like Givewell, which provides analysis that doesn't benefit itself, so it is a largely trusted broker.)

I'd be very interested in thinking about what other general strategies could exist - they seems like great targets for world optimization.

Oh boy I love this. Quite obvious, but such elegant analysis!

I have one example on top of this:

There is a phenomenon called “teal organizations”, where they replace top-down management with trust. Everyone in the company gets to decide which actions to take, as he’s probably the person closest to the problem. This is counteracted by additional responsibility to always ask someone more experienced than you when you make a decision — but instead of waiting until he gives you orders, you ask him yourself. Also, it doesn’t have to be your boss, it can be more meritorious colleague.

Sorry for the mixed description, you can read more at Reinventing Organizations by Frederic Laloux or at their website. There are many examples of successful organizations that works like that: Zappos, Valve are ones of the popular.

Overall, I love the series. Very deep and to the point, plus a lot of examples.

I think one additional area you can plug in here is regulation and federal business/commercial law as well. When that is done well it really should reduce coordination problems and relax constraints.

However, I thin one can says relaxing the coordination constraints is (probably) a necessary condition for increasing rewards at the object/skill level it doesn't seem to be a sufficient condition.

Still, as ALexei says, I do like you posts and make an effort to pay attention when I see one posted.

Good points. Personal to Prison Gangs makes a similar point about regulation, along with several other phenomena - litigation, credentialism, tribalism, etc. Based on the model in the OP, all of these are increasing over time because they solve coordination problems more scalably than old systems (e.g. personal reputation).

With regards to coordination vs object-level skills, I think a decent approximation is that object-level skills usually need to be satisficed - one needs to produce a good-enough product/service. After that, it's mainly about coordination: finding the people who already need the good-enough product you have. To put it differently, decreasing marginal returns usually seem to kick in much earlier in object-level investments than in coordination investments.

More generally, would it be reasonable to say that legibility relaxes coordination constantraints? That feels like it helps crystallise something I've been mulling over.

If you have a map of a city, you don't need to find someone who knows the city to navigate it. If you have accurate birth and death records for a country, you don't need to talk to local leaders to find out things like "how many people are avoiding taxes" or "how many people are dying of malaria". If you know exactly what quality of wood a forest is going to produce, you don't need to talk to all of the forest managers to find out which one is most suitable as a supplier.

Interesting point. Rings true, though I'll have to mull it over a bit. Certainly important-if-true.

The thought about the margins for object level and coordination level seems right. Perhaps another aspect to consider in that context is sources of increasing returns. The object level seems much more like the internal economies of scale. The coordination space seem much more like Smithian external economies/network type effects (which suggests my initial view was in error).

That's a great takeaway! Coordination is indeed an almost universal bottleneck.

Of course, just because decreasing coordination costs is good in general doesn't mean it's good for you in particular. There may well be somebody in India who can do your job better than you for a fifth the pay (telecommuting, probably), and it would be disastrous for you if someone solved your boss's coordination problem with that person. Trump campaigned on a platform of erecting barriers to foreign competition, and his election says quite a bit about how many people dearly wanted such a thing.