The Murmuration Strategy

The collective noun for a flock of starlings

A couple of years ago, Jason Kottke posted a video of a couple folks in a canoe who recorded a murmuration of starlings flying around an island.That was the first I learned the word “murmuration” and it’s stuck in my head ever since.

It’s difficult, and therefore especially important, for me to write about The Murmuration Strategy, as the principle behind it is just as fundamental to Rob’s Dream Company as the principle of hiring people with unrealized potential. It touches everything I want to do.

I expect to revisit this principle repeatedly, which really is the point of this every weekday exercise.

A murmuration — almost all flocking I think, but I like ‘murmuration’ as a word more than I do ‘flocking’ — features a very large number of individuals acting independently with influence by other individuals in close proximity, creating a large, beautiful, ever-changing image.

In the case of Rob’s Dream Company, these individuals are both employees and products.

As a good starting point, you can think of the public information about Valve’s reported ease of project switching. In my experience, working on the same project leads to stagnation, both for the employees involved and the project itself. By changing the team somewhat frequently, new life is repeatedly breathed into a project and employees are re-engaged on a regular basis with their work.

Murmurations lack a central co-ordinator. There is no one looking down to manage the swoops and whirls, the segments breaking off to do their own thing and the clouds of individuals merging into larger groups. I’m still undecided into how that is represented inside of a corporation, whether there is a hard cap on how many people and products can flow freely in this way.

I have little insight into other organizations who have tried similar strategies. Valve appears to, Google has done so (rumours about the removal of 20% time and the corporate-wide focus on Plus suggest they are moving away from it, though), 37signals appears to advocate for similar rotation of people and projects. I’m sure there are others patterning themselves after these successes and I’m unsure of how to create a culture in which this works.

My starting point is with products. My preference has been towards small, consumer-facing applications. I tend to not have a strong platform bias, though the one I have favours the web above all others. It’s what I know best, but I also like the potential for broad distribution, the technologies in play, and many people in the community. None of this implies the inverse for other platforms, as I like many things about each of them as well.

My goal is to release a small product, then another, then another, as time and money permit. Some of those products will, to push an analogy, fall right out of the next. Others will — I hope — join the murmuration, which will grow and change.