#24. Gall’s Law
A lot of my career has been focused on bringing new things into the world. In all the reading and learning that I’ve done over the last 15 years there is one short, profound paragraph that has embedded itself deep in my worldview:
“A complex system that works is invariably found to have evolved from a simple system that worked. The inverse proposition also appears to be true: A complex system designed from scratch never works and cannot be made to work. You have to start over, beginning with a working simple system.”
I’ve been thinking a lot about Gall’s Law this week. It feels like a good time to try and capture my thoughts about why it’s so foundational to me in making new things.
So let’s give it a go.
We swim in complex systems
The mundane truth is that we are surrounded by complex systems. Enmeshed in them. Pretty much anything that makes up our daily life - food supply, PAYE taxes, internet wifi hardware, organisations’ hiring processes, social media networks, rave culture and club promotion, and on and on - is a complex system. Anyone who’s read Donella Meadows will recognise the prevalence of complex systems in our world.
Gall’s Law says that none of these sprang into being fully formed. All of them evolved from simple systems that worked. But we don’t see this. And that costs us.
We don’t see it because - as Daniel Kahneman tells us - What You See Is All There Is. And what we see is complex systems! Unless you are paying deep attention to the history of system evolution for everything - impossible given how many systems there are and how difficult each one is to comprehend - then you’re blind that the working complexity you see everyday evolved from simple working systems in the past.
We dream complex dreams
Fine. This is just another cognitive bias. Until you try and make something new.
When we start thinking about making something new we tend to model the new thing on what we can already see around us. And what we see is complex systems! This leads us to dream of creating new complex systems without even knowing we’re doing it. Anyone who’s imagined a different future knows how easy it is to “see” our dream system in our mind. It feels so clear to us how it will work, how this feature will be useful, how this this risk-mitigation will be essential. It feels so real.
With this clear image in mind we unwittingly set out to design and deliver a complex system. We invest time and effort in thinking through all the facets of the dream system - the brand, the support model, the edge case features, the safety measures, the financial model, the internationalisation, and on and on and on.
Complex dreams fall apart
But this is a trap. And it’s a trap that limits our dreams.
Every new facet that we add to our dream system makes it more complex. Every change we make in the dream system has ripple effects throughout the whole. The work takes longer and costs more - and in taking longer new actors enter and add more facets with their own ripple effects. Before we know it our dream system - which once seemed tangible and real - is compromised, over-budget and long overdue.
Worst of all, not only has it taken ages, but when we launch it Gall’s Law tells us that it won’t work. That we might even have to start over, beginning with a simple system.
The unspoken tragedy of this is that the biggest dreams - the kind of dreams we need most in our society today - are exactly the dreams that are the most susceptible to these compromised, overdue, over-budget failures. The bigger the dream, the more likely you are to end up designing a complex system that simply does not work. And in the process you often burn your financial, political and social capital too. Argh.
Time is on your side
The answer isn’t to stop having big dreams.
I think that the answer is that once you have a big dream you need to step back and remember Gall’s Law. Specifically, you need to clear your head and figure out a pathway that starts with a simple system that works. And then put your trust in the evolution of that simple system.
It’s hard to trust in that evolution though.
One reason we find it hard to trust in evolution is because we’re bad at “seeing” time. When Darwin suggested that humans (complex systems!) evolved over thousands of generations from earlier life forms (simple systems!) many Victorians found it preposterous because they couldn’t visualise the ‘deep time’ of geology. In the same way, many 21st century knowledge workers can’t visualise the ‘rapid time’ of iteration and so find it preposterous that goals will be achieved in an unknown future.
Another reason we find it hard to trust in evolution is that it doesn’t feel intentional enough. Too random. But the evolution of a simple system doesn’t have to mimic the random mutations of biological evolution. Every time we learn about our simple working system - from observing usage, gathering data, talking with colleagues - we equip ourselves to evolve it in ways that lead towards our complex dream.
A simple working system plus time and guided evolution is radically powerful.
Gall’s Law runs deep
I spend my life trying to make new things. My fundamental role on teams is to plot a path to the complex dream we have for the future but that - somehow - starts with a simple working system. This is why Gall’s Law resonates so much for me.
So much of the world pushes back on this. We want our complex dream and we want it NOW! We are too sophisticated for this simplicity! In large organisations each functional discipline brings a default wishlist that creates high complexity before we even get started. Paradoxically, it’s often harder to get agreement on a simple working system than it is to get buy-in for the all-singing, all-dancing complex version.
As an aside, you might be thinking that Galls’s Law is just a rehash of Agile or Lean. I think it’s the other way around. Yes, Gall’s Law is part of what makes Agile and Lean powerful. But, for me at least, it’s a more fundamental philosophy that applies way beyond software development. Also it was written in 1975 :)
I’m sure there are domains and instances where Gall’s Law doesn’t apply. In fact, I don’t actually think it’s a ‘law’ at all. For me it’s just a powerful thinking tool when you’re trying to work out how to make a complex dream come true.
And, for me, that IS the work.
Originally posted on Substack along with Fleishman Is In Trouble, Luke Una, Perdido Street Station, birthday feast. Say hello or ask questions on @myddelton.