Conway's Law

How Your Software Team Structure is Shaped by Money

Ian and I first met at Container Solutions back in 2019 when he sought out a new challenge. His years in the gambling and finance industries gave him hard-won experience of running complex infrastructure under pressure, and that experience informed his ‘Hard Way’ series: Learn Git the Hard Way, Learn Bash the Hard Way and Learn Terraform the Hard Way. I made him an offer he could not refuse. Half the money and twice the work. He said yes.

Since then, we sparred continuously, with each other and our customers, about how best to rearrange teams so they can succeed with cloud computing and more recently with generative AI. These ideas collided in Ian’s mind last year when he made good on a threat he first made a decade ago. The result is Follow the Money, which I have the privilege of reading as Ian writes it.

I am going to explain Follow the Money to you now but I can only do that by explaining the chain of ideas on which it is built. The chain starts, like the best chains in the history of computing, in 1930s New York when that rotter Hitler chased Europe’s brightest minds out for, in some cases, practising the jüdische Wissenschaft. This explains why, when a young, local lad returned home from studying monkeys in Wisconsin, he found, much to his amazement, a near miraculous coming together of talent that had transformed his home town into the centre of the psychological universe.

Within the newly filled psychological cauldron of New York, the ideas of Alfred Adler, Erich Fromm and Karen Horney, all psychoanalysts, and Kurt Goldstein, Max Wertheimer and Kurt Koffka, all Gestalt psychologists, simmered alongside those of social scientists who had been given a safe home not at any of the Ivy league schools but rather at the New School that, desperate for funding and credibility, was happy to welcome them all. Abraham Maslow, aged only 27, stepped into the centre of this universe where he not only made friends and learnt from this wide range of mentors but threw a few ingredients into the cauldron whilst he was at it. After almost a decade of cooking, Maslow eventually brought from the kitchen his 1943 paper, ‘A Theory of Human Motivation’.

A Theory of Human Motivation turned the world’s understanding of motivation on its head. Previously, as Maslow later wrote,

The Stoics, most hedonists, practically all theologians, many political philosophers and most economic theorists have united in affirming the fact that good or happiness or pleasure is essentially the consequence of amelioration of this unpleasant state of affairs of wanting, of desiring, of needing.

Before Maslow’s paper, this ‘needs as a nuisance’ school of thought dominated thinking and taught that needs and the annoying, irritating motivation states they create, are to be done away with, got rid of, slain and splashed to the winds of heaven. Motivation, this school of thought would have us believe, is about reducing tension, anxiety and controlling human drives. With just a tiny prick of logic, Maslow popped this balloon. The need for love, for information, for safety and food might bring forth psychic troublemakers that do need attending to if the human is to avoid loneliness, gossip, the cold and starvation. These needs and, more importantly, the deficiency state that the human is put in if these needs aren’t met, is terrifying and rightly something to dread. But not all needs are to be dreaded. The higher level needs of humans, such as the need to move, the need to laugh, the need to educate ourselves and the need to create, give rise to pleasurable tensions. We don’t try to ameliorate them. We welcome them. The painter’s easel, the writer’s desk and the programmer’s computer draw us to them with a call that’s impossible to resist. The painter must paint, the writer must write and, much to their mum’s annoyance, the programmer must program.

The ‘needs as a nuisance’ school applies only to low-level needs for food, shelter and even love or what Maslow called belonging. It has nothing to say about higher level needs. But Maslow did. He showed that higher level needs have a different quality to lower level needs. Their gratification doesn’t decrease but instead increases motivation. Excitement is heightened rather than lessened and the person’s appetite intensifies. In short, us greedy humans want more. More education. More time to paint. More time in the garden. More time to explore. More time at the machine. As we gratify such needs, we don’t come to rest but rather we come to activity. The more a person gets, the more they want, which means this kind of wanting will never be satisfied and this gives rise to never-ending motivation.

Maslow’s triumph was to show that satisfaction of some needs doesn’t lead to decreases in motivation but to increases. He therefore turned the ‘needs as a nuisance’ school of thought on its head. By unravelling this mystery, Maslow planted the seeds of modern management techniques and offered an explanation as to why teams like mine, building web-scale applications and later doing so with the cloud, never, ever ran out of steam. And that brings us to Melvin Conway’s Law and Ian.

As an industry, or as managers learning our trade, we move from Maslow to Conway. Conway’s Law states that organisations create systems that mirror their communication structures. Conway’s Law therefore helps us to instantly see the link between cloud computing and management, and why you cannot be an expert in the former without being an expert in the latter. If managers want optimal systems, and optimal workflows where people are free to unleash their unlimited motivation, they have to organise their teams optimally, and reorganisation is always the job of management.

Thus in Maslow we have our why and in Conway we have the what: if managers want decent systems and to plunge those who build them into that dream-like flow state of unlimited motivation, they need to rearrange their organisation and its communication structures.

From the what of Conway we soon arrive at the how of Matthew Skelton and Manuel Pais. Team Topologies is an approach to structuring teams so they are optimised for fast flow. This is another way of saying that Team Topologies explains how to organise engineers, and other creative workers, so they are free to tap into their unlimited motivation and get some work done.

The problem with Team Topologies, however, is that managers cannot restructure for fast flow without changing the way money flows. That’s where Follow the Money comes in. Ian builds on Conway’s Law that by asking two questions. Where do communication structures come from and how can they be altered? The answer, it turns out, is money.

The way an organisation is funded determines its structure. Its structure determines its communication patterns. Its communication patterns determine its systems. Software architecture, in other words, is downstream of financial architecture, which means Conway’s Law is built on what I am now calling Miell’s Law:

Organisations that design systems are constrained to produce systems that reflect the financial structures and incentives of those organisations.

Follow the Money gives engineers and managers the tools they need to understand why their organisations are the way they are, and what it would actually take to change them. It is, in other words, his way of helping managers make the changes they need to optimise for fast flow, à la Team Topologies, which means Ian is really roping CFOs and hard-nosed executives into reorganising their companies along humanistic lines for the sake of both profit and sanity.

How does he do that? Well, partially by hammering the reader with some key concepts. The first is that money flows shape systems. Most engineers and managers know that communication structures shape software architecture because they understand Conway’s Law. Ian takes this one step further by showing that communications structures emerge from who funds what, on what cycle and on what terms. For example, centralised platform teams can and do create awesome shared services. But the consumers of those services don’t want to pay and thus the platform withers on the vine. Spending decisions, each down to a single manager, form (or deform) platforms and their teams. Thus, like following a river network shaped by the water to its source, we must follow the money upstream, and then redirect it to shape the downstream organisation which in turn will shape software and IT infrastructure.

The second concept is that of the patron. Patrons control the flow of money or value into a business. A patron might be a customer, a venture capitalist, a government regulator, a single founding customer or a private equity firm. Understanding who your patrons are, and what they want, explains more about your software architecture than any technical decision engineers ever make. For example, I once worked with a travel company whose board came entirely from traditional media. They organised their software development cycle like a print run, where a brochure is designed and approved before going off to print. In a time of A/B testing, that is not how web-apps are built. But the board decided what got spent and where and so they gave all the money to the manager who proposed a software delivery lifecycle that mirrored the old printing process. That experience highlights that when we follow the money, we find not just the patron but their mental model, too.

The third concept is scar tissue. Organisations accumulate financial structures that were once rational but are no longer. For example, teams exist because something went wrong five years ago or budget lines predate the products they fund. These ossifications remain unseen until you follow the money. One company we helped had a team that managed Elastic Search. We followed the money to their monthly wage bill and then told them that Elastic Search, newly available as a service, cost about thirty five thousand dollars less than that. They applied Conway’s Law and blasted that scar tissue away.

Wrap up

What started as a near-miraculous coming together of psychologists and social scientists ended up as Ian’s Follow the Money, built on the observation that money flows shape communication structure which in turn, as Conway taught us, shapes the systems that organisations build. Follow the Money rests on one more observation, the key theme of Visionaries, Rebels and Machines: structure, like that of a poem, provides freedom. Within that freedom self-actualisers fall into that dream-like flow state of unlimited motivation. The secret to this dream-like state is shaping money flows because they shape structure which in turn not only shapes the systems we build but also the systems with which we build them. That is Container Solutions’ secret, the worst kept in the industry but one that many companies still struggle with.

Where to now?

Grab a few beta chapters of Ian’s Follow the Money. The New School was welcoming in the same way the Ivy League schools where not. Eugenics, an Anglo-American project to its core, got packaged up by Madison Grant for Hitler to run with. In the same years, and not unrelated to this, antisemitism took hold of the Ivy League and that story is covered masterfully in Mark Oppenheimer’s ‘Gatecrashers’ podcast.

Grab the book