Because coordination is expensive
8 points by typesanitizer
8 points by typesanitizer
I think this article could benefit from an explicit answer to “why is coordination necessary?” After all, if coordination is as expensive as the article says, why shouldn’t some teams be exempt from it, or every team do 90% less of it?
A partial answer is that in some cases, spending time coordinating to agree on a solution takes less time than a team would spend on building the wrong thing because they haven’t coordinated. And in some cases, the bug that a team shipped (e.g. now that customers can enter 9-digit ZIP codes, the shipping label printer won’t print anything) or wrong action that a team took (e.g. an old database was deleted to save resources, but it’s actually still used) harms the company more than it saves coordination salaries.
That doesn’t explain how to decide how much time each organizational unit should dedicate to coordination. Maybe the linked Google and Zappos flat-hierarchy case studies have some answers.
Because if you were not coordinating, you would not be in an organisation, you would be a separate one!
I know this sound like a quip back, but it really is not. If you can avoid coordinating so much, then you are probably out of sync with the rest of the org in term of where you go, why, or what you are doing. And that may be a good thing. But what it does is separate you from the org to the point that you are probably going to be working at cross purpose with the org...
Or simply become an independent org. And sometimes, a bunch of orgs loosely related is a good thing. But the organisation, in order to stay one organisation, will naturally try to make you coordinate as a kind of survival instinct. Because if you do not need to coordinate, you are now offshoot of the organisation.
When was the last time you talked about your participation in meetings in your annual performance review?
I was in fact once asked by an organisation for feedback on a person's participation in a bunch of painfully-coordination-only meetings in an open-source project, because it was in fact a part of a performance review!
Hmmm, I am in a situation where performance reviews are somewhat optional, but once I get some unacceleratable parts far enough along (so that spending effort on this has any chance to make sense), I will need to write about a bunch of meetings to fill one of the facets of review…
It's another story how in a chaotic enough org the meetings themselves can de-facto become shadow performance review.