On Feedback Loops
I’ve been noodling a fair bit lately on organizational design, team dynamics, and work processes. How does it all work? In part, I think, cultural feedback loops.
A feedback loop is the phenomenon where a system’s outputs are fed into its inputs. In audio contexts, it’s generally a moment of pain (unless you’re Jimi Hendrix), but in a lot of systems, it’s neither good nor bad, it’s just a characteristic of the system.
Here I want to think about cultural feedback loops: i.e. various scenarios when people participate in feedback loops with other people. I’m going to suggest that it turns out we spend much of our time in feedback loops. In effect, feedback loops form our social fabric, they create culture.
Let’s start with the simple case of two people in some kind of relationship, be it business or personal. In Games People Play, Eric Berne explores the notion that we often engage in patterns that amount to a game, complete with tacit rules and so forth.
Part of the strength of the book is the catalog of different games he provides. For example, one of the games he describes is the Why Don’t You — Yes But game. It describes two people, let’s call them The Complainer and The Problem Solver, who engage in a perpetual feedback loop of complaints and then solutions, and then yes-but’s extending the complaint, and then more solutions, and so on and so on.
Berne’s examples are almost caricatures, but at the same time they lay out recognizable patterns. I hope that one example resonates somewhat. He has dozens more to leaf through. (Although, full disclosure: the book was written in 1964 and some of the examples and wording certainly speak to the Mad Men era it was written in.)
There are a couple interesting qualities to these games: first and foremost, they are self-reinforcing and self-correcting. If one of the interlocutors were to stop their side of the feedback loop, the conversation would grind to a halt. To prevent this, the second interlocutor will steer the conversation back onto the established pattern. Second, the pattern isn’t really named (except by Berne) or identified. It’s what we do, it’s not necessarily what we say we do.
Stepping away from Berne’s archetypes, real-world interactions often have more than one set of transactions in play at the same time. And real-world interactions also often have more than two participants. But we can take the same concept and generalize it to team or organization interactions. The same basic feedback loop principles apply, but in these larger settings, there are probably multiple feedback loops working in tandem, or multiple different patterns forming the different arcs that comprise the full circle of the feedback loop.
So the three traits of social or team feedback loops are:
- they are self-reinforcing and self-correcting
- they are generally unnamed, or named only indirectly
- they can be decomposed into component transactions
My real point is here is that on a team one both does and doesn’t have a choice about their feedback loops. On the one hand, you don’t have a choice due to the first two traits above: absent any disruption, a feedback loop will self-reinforce and, if operating effectively, will go unnoticed. This can be a very wonderful thing when it is Just Working, and probably accounts for all sorts of efficiencies and comity amongst the team.
At the same time, there are choices in the sense that there are always multiple alternate feedback loops, and each has its merits and its flaws. And due to that self-correcting property of feedback loops, it will take some kind of disruption to alter the course of a feedback loop. The way to a controlled shift to a new feedback loop will start with decomposing it into its component transactions, giving voice to them, and choosing to break or to transform those transactions.
My takeaway here is that a healthy organization should be vigilant about its feedback loops, and be attentive to whether those feedback loops could use some disruption or adjustment. And last, when disruption is the right answer, it’s wise to be aware of the self-reinforcing strength of a feedback loop.