Skip to content

July 15, 2022

On Coordination Costs: Moving A Couch, And Painting A Room

By Gene Kim

I recently posted an essay “On Coordination Costs and Organizational Architecture: Book Thesis”, which describes the thesis of the book I’m working on with Dr. Steve Spear.  We spent the week together in Boston two weeks ago, and I was so energized by the following essay that describes the problem of coordination costs for a very simple problem.

Does this resonate with you?  Please leave a comment with any thoughts!

Even Moving A Couch Requires Brain Work

What we have observed is that across every activity across every industry vertical, high performers have figured out how to vastly reduce the cost of coordination.  

To better understand coordination costs, consider doing something as basic as moving a couch with a friend — let’s say their names are Steve and Gene.  When they start, they are immediately in conversation, actively communicating and coordinating: where do you put your hands, how do you keep the couch balanced, do you move the couch sideways or lengthwise?

To fit the couch through that narrow doorway, do we need to lift one end of the couch vertically, and if so, whom?  To get down the stairs, who goes first, and do they go down the stairs facing backwards or forwards?  

Through actions, trial and error (i.e., feedback), conversation, and correction, we can be confident that Steve and Gene will eventually figure out how to successfully move the couch.

Now suppose that the homeowner shows up and keeps butting in, imposing their strong opinions on where they should hold the couch, who should go through the door first, and so forth.

Worse, maybe this person starts interrupting Steve and Gene from communicating directly, or even insists that all messages be relayed through them.  Suddenly, urgent messages such as “the couch is slipping, can we slow down?” or “the couch is pinching my finger in the doorway” are no longer making it through to the other person.

The outcome is almost assuredly that the time and effort required to move the couch goes up — and the work becomes far more dangerous, as well.

This is because we are forcing information to go through an intermediary, dramatically increasing the cost of coordination, and making it far more difficult for Steve and Gene to get their job done — we lose information, we introduce delays and potential misunderstandings, and in short, get in the way of the people solving problems and doing their work.

The effects of coordination costs is evident in even moving a couch, because it requires coordination between people — it can’t be done by automatons, requiring human creativity and problem solving.  It is not just “brawn problem,” but a considerable “brain problem” as well, and requires considerable information exchange for the work to happen well.

And if coordination costs can dramatically affect the outcomes of moving a couch, consider how corrosive it can be to building a plane, train, automobile, or sending someone to the Moon and safely returning them to the Earth, or building and operating the software systems that enable all of those activities.

Extending the Couch Moving Scenario

Let us revisit the scenario of Steve and Gene moving the couch — they need to move the couch because their wives, Miriam and Margueritte, are painting the walls and ceiling of that room, and want to make sure that paint doesn’t get on the couch.

Now in the room are the two people moving the couch, as well as two painters.  Also strewn across the room are open paint cans and four ladders.

We have now increased substantially the amount of required coordination — Steve and Gene need to be careful not to kick over the paint cans, or bump into the ladders or their wives. And Miriam and Margueritte have to be aware of the locations of Steve, Gene and the couch, so as not to drip paint on them.

As a result, everyone has to communicate what they want to do, where they want to go, perhaps ask that things be moved, and jostle and rearrange work to accommodate the other people’s needs.  We may even have a problem with dependencies: Steve can’t move his end of the couch until Margueritte moves her ladder, but she can’t move her ladder until Miriam is finished painting her section of the wall.

And despite all the efforts to coordinate, mistakes can still happen — paint cans can be spilled onto the floor, ladders can be tipped over (potentially with the painters on it), paint can drip from the ceilings onto the couch, and so forth.

And even if none of these things happen, we are now spending more time and effort coordinating: painters can’t focus on painting, because they’re focused on the couch.  And  couch movers can’t focus on moving the couch, because they’re worried about the paint.

The result is that we have a lousy painting job, and maybe divots in the wall caused by accidents moving the couch.  

Two Ways To Partition To Create Modularity

We propose that there are two ways to reduce coordination cost — 

The first method is to partition the activities by time: for example, Steve and Gene move the couch at 10am, and Miriam and Margueritte paint at 11am, when the couch has been safely moved out of the room. And when they are done painting at the end of the day, Steve and Gene bring the couch back into the room.  

By doing this, the couch movers and painters are not in the room at the same time, so their actions no longer affect each other.  By merely agreeing on the separate times and places where their work will be done, the teams can work independently of each other. And astonishingly, they are now working in a more coordinated fashion.

Coordination requires communication, which is the exchange of information — and by partitioning the two activities in time, we have radically reduced the amount of communication needed and information that has to be exchanged. 

By doing this, we dramatically reduce the amount of energy spent on coordination, which can instead be used to solve the real problems in front of us. Furthermore, we reduce distractions and interruptions, which increases our ability to focus. Studies have shown how critical focus is for cognitive tasks — for instance, when we are driving at night in a snowstorm, we will slow down, turn off the podcast, and ask the kids to be quiet, so we can concentrate.

The result is that we get a better paint job, and the couch is moved without damaging anything, and it was all be faster, easier, and less dangerous to do.  And by removing distractions from painting, even the most intricate and difficult parts, such as painting the corners of the ceiling, can be done with focus and precision.

We have figured out how to harmoniously integrate the work of moving the couch and painting the room.

On the surface, solving the problem of moving a couch to paint a room to reduce coordination cost may seem laughably simple — however, this is a microcosm of what all high performers are doing across nearly every industry vertical, but at a much larger scale.

Other Method: Partition By Space

Incidentally, when we cannot partition the activities by time, we can partition the activities by space — in this scenario, the couch movers and painters will work opposite sides of the room at the same time. For instance, Steve and Gene will move the couch on the left side of the room, while Miriam and Margueritte paint the right side of the room.


When the right side of the room has been painted, the painting and couch will switch sides.
As in the previous example, by partitioning the activities, coordination between teams is needed only during the transition times — the couch movers no longer have to worry that moving the couch will affect the painting, and vice versa.


The outcome is the same: a far better painting job, with less time and energy required to coordinate.

Upcoming Book

(These are topics which will be part of our upcoming book — if you are interested in staying up to date on the book’s progress, which currently does not even have a working title, but will be released in 2023, please sign up here: https://itrevolution.com/kim-spear-book)

- About The Authors
Avatar photo

Gene Kim

Gene Kim is a Wall Street Journal bestselling author, researcher, and multiple award-winning CTO. He has been studying high-performing technology organizations since 1999 and was the founder and CTO of Tripwire for 13 years. He is the author of six books, The Unicorn Project (2019), and co-author of the Shingo Publication Award winning Accelerate (2018), The DevOps Handbook (2016), and The Phoenix Project (2013). Since 2014, he has been the founder and organizer of DevOps Enterprise Summit, studying the technology transformations of large, complex organizations.

Follow Gene on Social Media

7 Comments

  • Pingback: PRODUCTHEAD: What do you worry about on a Sunday night? – I Manage Products

  • Steve Spear Sep 2, 2022 12:38 am

    Gene. This is super. And cut and paste right into the book text, I think. But seriously, right on an expression of the points so good.

  • Craig Statham Aug 2, 2022 8:16 pm

    If we partition by time then the question becomes what are Miriam and Margueritte doing while Gene and Steve move the couch? If we can't give them other work to do (outside of the room) then their time is lost which becomes a cost. How long will it take for the painters to become productive after Gene and Steve have performed their task? Is there hidden cost in this kind of hand-off? Safely partitioning the space with high frequency updates between the two teams to co-ordinate when it's time to move seems more effective. They work locally in isolation of each other but understand the intent of the collaboration to get the job completed safely in less time.

    • Nicole Sep 10, 2022 8:43 am

      … then their time is lost which becomes a cost. When I hear this phrase, I always think of the highway example. When it's at 100% capacity, nothing moves and no one makes any progress. I wonder if that doesn't also apply to our work. We do one task after the other in a completely structured daily routine...if something goes wrong, our whole house of cards collapses. I think we need a discussion again whether 80% utilization is not rather the right measure. You're right, of course, the costs arise, even those who do nothing. But this "free time" can significantly reduce the effort for coordination.

  • Kent Craig Aug 2, 2022 7:25 pm

    Communication, Collaboration, and Coordination - the better you do these three things, the better your chance of success. It is not a matter of more of these three things, but of efficiency and balance. Love your books, Gene! The demise of Sarahs.

  • Tom Catalini Jul 25, 2022 12:36 pm

    Love the phrase "coordination costs" which captures the concept well without making it sound to jargony. I wonder about "tracking costs" driven by the practical needs for each person to be aware of status of the other's activities, as this is needed for each coordination method. But also for reporting up to the homeowner or other stakeholders...which maybe begs the question of what is necessary in that scenario. Maybe nothing. But maybe it's critical if other projects are happening in the house, which is the typical case in most environments.

  • Dinkar Gupta Jul 15, 2022 8:05 pm

    First all, very real scenario for me.. have been there, done that. We also had another party to the mix, a contractor fixing the floor tiles.. so it was a 3 dimensional chaos :-) Two thoughts for consideration. for partition by time, task completion commitments have to be considered, which mean ordering comes in. so.. are we shifting the "in-task" coordination cost to "pre-task" coordination ? Is this really reducing time in holistinc sense or local optimization during work. maybe a good thing as pre-defined agreements are not a bad idea. Its also a matter of trade-off I guess and also availability of both parties to at those times. Do number of coordinating parties help decide whether "pre task" coordination is costly or "in task".. aka shifting from runtime to design time. Plus the idle time (in the context, while other team is working in room.. assumption is they are producitvely using the time. May not be. (OS process and task scheduling scenarios similarity) Partition by space, there number of possibilities exist. However here the determinant is the space itself. Another point.. Does task decomposition play a role here ? I think that's another aspect worth looking at. Can the couch be moved in parts ? Is the paint job a 2 step process where one can be done with couch movement while other not.. so product modularity and process design ? One last bit. "Cost" .. I am assuming we are talking about cognitive and physical effort primarily. However what if material cost & properties play a role.. what if the paint can't wait for long and may degrade if waiting too long after mixing.

Leave a Comment

Your email address will not be published.



Jump to Section

    More Like This

    Attack of the Supply Chains – Investments Unlimited Series: Chapter 9
    By IT Revolution , Helen Beal , Bill Bensing , Jason Cox , Michael Edenzon , Dr. Tapabrata "Topo" Pal , Caleb Queern , John Rzeszotarski , Andres Vega , John Willis

    Welcome to the ninth installment of IT Revolution’s series based on the book Investments…

    Finding Digital Treasure Through the Science of Happy Accidents
    By Summary by IT Revolution

    In a recent talk at DevOps Enterprise Summit 2023, industry veterans Steven Fishman and…

    Lessons from Unexpected Places: How Watching Ted Lasso Makes Us Better at Agile and DevOps
    By Anthony Earl , Jordan Stoner

    It happened to us, and we bet it’s happened to you too—you watch a…

    The Three Lines Model: Investments Unlimited Series: Chapter 8
    By IT Revolution , Helen Beal , Bill Bensing , Jason Cox , Michael Edenzon , Dr. Tapabrata "Topo" Pal , Caleb Queern , John Rzeszotarski , Andres Vega , John Willis

    Welcome to the eighth installment of IT Revolution’s series based on the book Investments…