Product, organizations and the system lens: and what digital signal processing has to do with it
Paraphrasing Russell Ackoff, we "are not confronted with problems that are independent, but with dynamic situations that consist of complex systems of changing problems that interact with each other". He called such situations messes… Thus, we don't solve problems, but manage messes.
If you've been around reading this, or just know me well-enough, you should know by now that I find the system lens as one of the most useful perspectives to take. More often than not, a complex adaptive system is the most precise way to define what we deal with, on an ongoing basis.
From a product point of view, whenever we are trying to solve a problem by intervening somewhere with a change, there's a tangible chance that we might mess with something else. For example, we may put too much emphasis and make acquiring new users so easy, to then learn that most didn't last long enough to make a difference.
And if you don't have the feedback loops with suitable frequency, you may learn too late about the unintended introduced issue, to be able to do something about it timely.
Recently I read the latest book from Gene Kim (written together with Steven Spear): Wiring the Winning Organization. There I learned about a theorem that helps to explain the overarching idea of having suitable frequency to learn and do something about what has been sensed. It comes from digital signal processing and it's called the Nyquist-Shannon Sampling Theorem. Removing all the bells and whistles of the theorem in its original technicality, the underlying idea is easy to explain intuitively:
You can only effectively control a system if the frequency in which you are reading and perceiving the changes is twice faster than the rate of change. Put in simpler words, if all you can sense is an update given every other week, you are essentially accepting not being able to use that to control (manage) situations that can change more frequently than four weeks (twice of the sensing frequency).
Think about the possible deeper repercussions of that for a moment. And attempt to project that to the dynamic, often fast-pacing, context of product development we experience in a digital world. Imagine you have set some quarterly goals (maybe you used OKRs for that) to achieve a certain outcome or a related output metric that you believe to be linked to that. But you thought of not making the follow up of that too much of a burden, so you decide to go with a monthly check-in of progress.
By the principle articulated in Nyquist-Shannon Sample Theorem, you have to be willing to accept that you will only have one chance, the first check-in after the quarter started, to be able to sense a need for changing some direction, assuming the underlying rate of change is potentially faster.
As I thought further about that, it really blew my mind how often we are willing to settle to a much less frequency of controlling something in organizations. Whereas at the same time we seem to acknowledge that our fast-paced digital world requires the ability to be much nimbler and adaptive, in much higher frequency than what is sometimes used to make strategic calls for a change of direction.
Something doesn't quite add up there… So, no actual surprise on how often organizations will miss sensing some critical signals and to take meaningful actions in a timely manner, before it's likely too late. What tends to follow is an eventual need for a much more abrupt intervention (e.g., think of major reorganizations, which typically have to do with lack of good ideas, or missing the signals to act timely, or a combination of both: signals were sensed but there weren't good ideas available or sufficiently surfacing), because now things are tending too much towards chaos, and they are not managing their messes well-enough anymore*.
By Rodrigo Sperb, feel free to connect, I'm happy to engage and interact. If I can be of further utility to you or your organization in getting better at working with product development, I am available for part-time advisory, consulting or contract-based engagements.
* There's a caveat and there's such a thing as reacting too fast to signals (which are then more like noises than useful things to be sensed and acted upon. But I will leave it for another time for the sake of sticking to a point and not making this too long.