The Org Chart is Dead
It all started with Scottish-born engineer Daniel McCallum in 1855. McCallum had joined the New York and Erie Railroad as a subcontractor building bridges, and by 1854 had risen to general superintendent. The railroad had grown so large and complex that it was no longer clear who was responsible for acting on the information that could prevent train crashes. So, with his colleague and civil engineer George Holt Henshaw, he drew a picture of how the business was structured. The Org Chart was born.
McCallum’s image was a large tree. The people with the most authority sat at the bottom of the chart, as the roots, with departments and personnel, from superintendents and engineers to labourers, as the branches and leaves. His organic picture emphasised the importance of middle management.
The term ‘organisation chart’ only came into use in the early twentieth century. In 1914, Willard Brinton complained that engineers weren’t using Org Charts as widely as they should. But by the 1920s Org Charts had found their way into the business world.
At around the same time, French mining engineer Henri Fayol suggested that the military idea of “chain of command” was useful and the tree was replaced by a hierarchical pyramid. Fayol believed an employee should receive orders from only one superior, echoing the biblical teaching that “no man can serve two masters.”
So, the Org Cart’s ‘parents’ were the Bible and the Military. God and guns, quite the lineage.
One of Fayol’s fans was Lyndall Urwick, a Major in the First World War who won the Military Cross. After the war he joined his father’s glove-making business, Fownes Brothers, and became the first to apply the idea of “span of control” formally to business, He believed three to six direct reports was optimum.
In the early 1920s, Alfred Sloan, with General Motors’ chairman Pierre du Pont, restructured GM into a multidivisional operation. This was designed to decentralise decisions, but their Org Chart was a classic pyramid. For the next half-century or more, the Org Chart was how companies standardised their design, and it helped them grow to an unprecedented scale, particularly from the 1950s onwards.
By the late 1950s and 1960s, rigid pyramidal designs, inspired by the Military and the Bible, struggled to cope with an increasingly complicated world. A new approach started to emerge as America reached towards the heavens.
NASA and its defence contractors faced a problem the pyramid couldn’t solve: putting humans into space required integrating propulsion engineering, avionics, life support, materials science and dozens of other specialties simultaneously. A siloed pyramid was incapable of delivering that level of interdisciplinary connectivity. John Mee, an Indiana University management professor, watched this new form take shape and, in 1964, gave it a name: the “Matrix” organisation. It was baked into organisational folklore in 1977, when Harvard’s Stanley Davis and Paul Lawrence published Matrix, the defining book on the subject.
The trouble was that the Matrix broke Fayol’s oldest rule. Giving employees ‘hard’ and ‘dotted’ reporting lines meant asking them to serve two masters. Add the Matrix’s love of ‘double’ and ‘triple hatting’, and you had a recipe for chaos. But the most influential critique came in 1990 from Harvard’s Christopher Bartlett and INSEAD’s Sumantra Ghoshal. Their argument was less about the Matrix idea itself, and more that people living in Matrix organisations lacked the developmental sophistication to handle the complexity they were facing.
Meanwhile, from the early 1980s, a very different response to complexity was gathering pace: the new idea was delayering which later became rampant re-engineering. Its poster child was the youngest Chairman General Electric had ever seen, Jack Welch. Between 1980 and 1985, “Neutron Jack”, a nickname earned because he loved eliminating employees while leaving buildings intact, took GE from 411,000 employees to 299,000, through layoffs, sell-offs and the dismantling of nine layers of management.
Michael Hammer’s 1990 HBR article sensitively suggested “Don’t Automate, Obliterate”.
Hammer nailed his ideas with a book he co-authored with Jim Champy, called Reengineering the Corporation. Hammer coined the term reengineering. He had considered calling it transformation but rejected that word because it sounded too touchy-feely. Reengineering was hard, tough, masculine.
Hammer was fond of saying, “In reengineering, we carry the wounded and shoot the stragglers.”
The reengineering movement drew on Toyota’s lean manufacturing, popularised in the West by MIT’s The Machine That Changed the World (1990), and on Milton Friedman’s doctrine that the social responsibility of business is to increase its profits. Friedman sought to give cost cutting a moral licence. Many companies subsequently used re-engineering as a pretext for dramatic downsizing, and the term became synonymous with redundancies and layoffs.
By the mid-1990s the backlash against reengineering was so fierce that its three founding figures, Hammer, Thomas Davenport and Champy, all issued public apologies. In 1996, Hammer told The Wall Street Journal: “I wasn’t smart enough. I was reflecting my engineering background and was insufficiently appreciative of the human dimension.”
Wayne Cascio’s research through the 1990s found that downsized firms often failed to outperform their peers on profitability. And, as network research has since shown, one of the main reasons for this is indiscriminate cost cutting often meant a loss of institutional knowledge and, crucially, the informal networks that actually got work done.
In 2001, seventeen software developers gathered at a ski resort in Utah and wrote the Agile Manifesto. It advocated for small, self-organising teams, customer collaboration and the ability to respond quickly to change. It was humane and network-friendly in spirit. But like many things it morphed as it scaled up in the 2010s and became a consulting product with squads, tribes, chapters and guilds. Basically, a matrix with better branding.
In 2023, the axe returned wearing agile’s hoodie. Mark Zuckerberg declared Meta’s “Year of Efficiency” and attacked “managers managing managers”; Amazon, Google and Salesforce soon followed and the ‘Great Flattening’ appeared.
From 1855 to today, organisational design has gone through several metamorphoses: from the great pyramid, to the matrix with its extra lines and hats; to the delayering axe; to the chaos of “agile at scale” with fewer boxes; and most recently to the ‘Great Flattening’, which is basically the axe rebranded. Each generation redrew the boxes and lines. None of them asked how work actually gets done.
The Real Org Chart
As a result, most organisations’ Org Charts bear little or no relationship to how their company actually functions.
We know this because, using our Organisational Network Analysis (ONA) technology, we’ve studied more than 65 companies and explored more than 1.3 million relationship dynamics to determine how modern businesses really work, regardless of what their Org Chart says or what functional, regional, pyramidal, agile or matrix operating model they claim to be using.
Our data paints a profoundly different picture from the standard Org Chart, and from what most leaders think their organisation looks like.
Three-Layered Networks with a Semi-Permeable Membrane
What we see is a much more organic, amoeba-like, three-layered network of relationships, which looks profoundly different depending on whether you’re exploring operational, cultural or strategic connectivity.
Your networked organisation is often surrounded by a ‘semi-permeable membrane’ as you divest or ingest teams of capable individuals, or move from supplier relationships to partnerships with other players in your supply-chain ecosystem.
The ONA technology generates incredible insights into how work gets done and can precisely quantify the level of influence of every single person in the organisation, supported by hard metrics. These measures are based not on the opinion of a handful of leaders or line managers, but on the wisdom of your employee crowd. And they can tell you what’s really going on in your company much more accurately than a manager who wants to project a certain narrative.
ONA data insights have massive implications for talent pool management, succession planning, attrition analysis, cost management, risk analysis and change, as well as organisational and team performance.
More recently, we’re seeing that ONA data can be a game-changer for AI adoption programmes. The never-ending desire of corporations to reduce costs and increase profitability is making agentification seem much more attractive.
But many organisations don’t realise there are massive risks if they replace uniquely human capability with agents or remove people who bridge different parts of the system because they didn’t know these were the key links that make the whole thing work.
There are a few enlightened organisations, like Vodafone’s VOIS, that are actively building “shared intelligence”, integrating AI with uniquely human capability. But they are currently the exception.
If you want to see how your business is really organised, who your key players and high-potentials (HiPos) really are, and how to drive successful AI adoption, then reach out to us. We may just be able to save you a massive amount of suffering, not to mention time and money.