Squeezing the toothpaste tube
Complexity and org design
The King Of Wishful Thinking
They called it ‘the sponge’. The layer of senior and middle management between the board and the workface. The board would lay out strategy, give directions, issue mandates, all intended to change how the business operated on the ground. But the sponge absorbed it all, and everything below them carried on as usual.
It worked the other way too. Whatever unrest and dissatisfaction, whatever new ideas and initiatives were happening amongst the workforce, none of it percolated up to the board. The sponge soaked it all up.
The sponge was a major frustration to the board of BT in the late 1980s. What they didn’t understand is what the sponge actually did. It was a necessary buffer between the board and the workforce, reducing friction and conflict. It was an essential interpretive layer between the high-flown language of strategy and the realistic actions to be carried out on the ground. It evened out the ups and downs, it absorbed the everyday shocks, it was the lubricant that kept the machine moving.
Of course, there were some problems with the sponge. There were some managers who were resistant and insisted on doing their own thing. They would interpret any command from on high in a way that suited their interests, even if that was clearly contrary to the spirit of the command. They would be capable of mental acrobatics and linguistic dexterity such that they could claim black was white and keep a straight face. The CEO of my division was one of these and the logical contortions he was capable of made Alice in Wonderland seem like real life.
What the board should have done was identify and remove these managers. What they did instead, of course, was call in the consultants (I can’t swear to it but I think it was McKinsey’s).
So what we got was the largest organisational restructuring programme in Europe, Project Sovereign. The concept was simple. There would only be six layers of management between the board and the workface. Managers would have between 6 and 10 direct reports. This delayering would dry out the sponge and create a much more responsive organisation. Generous voluntary redundancy packages would ease the pain of shedding thousands of employees.
The new org charts were published and those of us that remained fitted into neat, flat structures. There was a period of confusion as we tried to figure out how to do our jobs in this new structure and how to plug the holes that had inevitably emerged. Then, as the fog cleared, something entirely predictable happened. The sponge began to grow again.
The new structure just didn’t work. Managers were either over-stretched because they had too many direct reports, or were underemployed because they now didn’t have any and had to wait for approval for decision the could previously take themselves. So gradually lots of ‘dotted lines’ began appearing on the org chart. A dotted line showed the day-to-day reporting line. Often this dotted line was to the person who used to be your boss but was now supposedly at the same level as you.
Over time, the dotted lines got thicker, because why would your notional ‘boss’ do your annual appraisal when they didn’t manage you day-to-day - and probably didn’t really know what you did or, in some extreme cases, who you were.
And so within about 18 months, the new structure had morphed into something that more closely resembled the old structure. Not because managers didn’t want the new structure (although there was no doubt some of that) but because it just didn’t work. Reality meant you need more than six levels of management. Life was more complex. Who knew?
Harder Than You Think
I was reminded of this by a great article Stefan N. wrote on his Synexia substack, “The Hierarchy of Work Complexity Is Inescapable (Even at Buurtzorg)”.
It was one of those times when you read something that causes a fog to lift and brings clarity to something that has been lurking, only part-formed, around the edges of your consciousness. A real ‘aha’ moment, when a truth is revealed that then looks so blindingly obvious you wonder why you couldn’t see it in the first place.
So what is this great insight? Well, let me use Stefan’s own words:
“What I understand now is almost too obvious to state: you cannot remove complexity; you can only choose where it lives and how to absorb it. And if you don’t make that choice consciously, it will find its own habitat anyway, usually somewhere dysfunctional and with a high cost.”
The way I interpret this is that in a conventional hierarchy, the complexity largely sits in the middle management layer. That’s what ‘the sponge’ does, it handles the complexity. Not always very well but well enough to stop the organisation being overwhelmed. What we’ve seen over the past few decades is that complexity has increased and so the load on middle management has risen. This has had two effects. The first is to make the role of the middle manager much harder (almost impossible, some might say) and that has led to higher levels of burnout.
The second is a big increase in middle management because that’s the only way a hierarchy can respond to increased complexity. (This is my interpretation, not Stefan’s!). There have been several attempts to slim down middle management, to delayer, but it just grows back again (as it did at BT after Project Sovereign).
Attempts to apply the self-management model of Buurtzorg and Haier elsewhere have been unsuccessful. This is because they have been applied superficially, simply removing management without relocating the complexity, all the while proclaiming “Hierarchy is dead!”. You can flatten your organisation but the complexity doesn’t go away, no matter how hard the board might wish it. It moves to the most unwelcome place instead.
Here’s Stefan observation:
‘The truth is that Buurtzorg doesn’t eliminate complexity at all. It simply relocates it into different structures: architecture, software, contracts, professional norms, and the macro-institutional environment of Dutch healthcare. And once you see where the complexity has moved, the whole model becomes not just plausible but elegant.’
(Buurtzorg, the dutch community healthcare provider, has nurse practitioners who work in teams of up to 12 and they self-manage. They are supported by a team of coaches, who they can call upon when they have problems; a small central support team that provide back office services; and an intranet for information, knowledge sharing, co-ordination and collaboration. It is significantly more effective than its competitors, both in terms of outcomes and costs.)
He identifies 4 different types of complexity - operational, co-ordination, integrative and environmental. ‘Buurtzorg’s achievement is not that it removed these loads, but that it redistributed them: operational and coordination complexity live inside the teams, while integrative and environmental complexity live in the architecture (e.g. the intranet) and the institutional environment. That’s the real design innovation.’
So simply taking Buurtzorg’s structure and plonking them down in a different country won’t work, because the complexity that sits in the Dutch healthcare system, for example, needs to find a new place to sit in a different institutional environment.
With Haier (the world-leading white goods producer from China that consists of thousands of micro-enterprises (typically under 15 people), operating within a broader ecosystem supported by a suite of networked applications) we see a different approach. ‘Haier uses internal markets, dynamic contracts, and competitive bidding to allocate and absorb complexity’
In both these case, they can reduce the hierarchy because they have moved the complexity elsewhere. That’s the ‘power move’ that makes it work.
They are not post-hierarchical, they are post-managerial.
(The above is my interpretation of this article and what I took away from it. If I have mangled it in the process or been superficial, I apologise. It is quite technical in places, and the discussion that it generated with Dr Richard Claydon, Otti Vogt and others gets deep into management theory and philosophy and I’m not sure I followed all of it! But if you’re into that, it’s well worth reading it all - you can pick up the trail here.)
P.S. I’ve just realised that this could be summarised as ‘organisational design is like squeezing a tube of toothpaste with the lid on. It doesn’t matter where you squeeze it, the toothpaste just moves somewhere else’. The toothpaste here being, in part at least, the complexity. The reason this occurred to me is that this is exactly how my bosses at BT described how that organisation failed to respond to the many reorganisations that they tried (and yes, it’s where this week’s title came from). Needless to say, none of their organisational wheezes took any account of the complexity, they just wished it away.
All The Small Things
What should also be noted is that these organisations are based on a few simple, yet powerfully elegant, guiding principles. They have a clearly articulated vision about what the organisation is about.
In the case of Buurtzorg, they had a clear idea of how neighbourhood care should be delivered.
In the case of Haier, their animating principle is ‘zero distance to customer’. Their micro-enterprises are directly interacting with their customers and empowered to respond to that customer demand quickly. A second principle is that ‘Everyone can be a CEO’, which firmly places a premium on enterprise and initiative and also embodies their hyper-capitalistic model.
To add another, Morning Star is a large US tomato processor that uses self-management through contract commitments that are negotiated between the individuals that handover to each other in the process. It has two guiding principles that everyone signs up to. They are:
Never use force
Honor your commitments
The organisational structures grow out of these principles, these philosophies, if you like. These organisations are successful not just because of the efficiencies their organisational structures produce but because they have a unifying purpose, what you could even call a credo.
These are profound perspectives. They are not arrived at easily or by chance and they certainly aren’t going to come in a consultant’s powerpoint.
Conventional hierarchical organisations are struggling to cope with the increasing level of complexity. It’s overwhelming its organisational home, middle management. The response is to increase process and control, which makes the organisation more rigid and less capable of managing complexity. Or to remove layers of management, which has the same effect. They are effectively wishing the complexity away, as evidenced by the increasing level of mandates to be in the office and have contact days in order to improve co-ordination, collaboration and ‘culture’ (but which has the opposite effect).
New models will emerge but a deeper level of thinking is needed than the ‘copy and paste’ approach of moving to a ‘flat’ organisation or implementing self-management (which can often become managerial abandonment).
Are the leaders we have in place now capable of that depth of thinking? I’m sceptical but even if they are, will they get the space and time in our current capitalist system? Will they be able to escape the pressure of stock market expectations for growth and profitability, the quarterly scrutiny of analysts, or the reward structures geared to short-term performance?
Hurry Up And Wait
To that point, recent research by Microsoft shows that an employee’s opinion of their manager is positively affected by the manager’s speed of response. If there is a delay in the manager’s responses, the employee rates them less highly.
This is reflective of a trend we have seen for decades, where speed of response is valued more highly than the quality of the response. As I noted two weeks ago, this began with the telephone, which made hesitation seem like indecision, even if it was actually someone being reflective and thoughtful before answering. It led to management valuing people who could ‘think on their feet’, judging them by their speed of response rather than the quality of their thinking. The fast but superficial took precedence over the deep and thoughtful, even though the former often led to more heat than illumination.
The arrival of email sped this up, and instant messaging accelerated it again. Social media has been another accelerant, emphasising not only speed but brevity. However, without the time and skills to craft these responses, meaning is lost and they become more shallow and superficial.
In a work environment that increasingly values busy-ness over effectiveness and presentation over substance, it seems inevitable that management should be measured by something that is not only pretty meaningless but is itself a sign of poor management and leadership.
We are prioritising speed of bullshit over depth and substance. And, yes, AI will probably make it a lot worse.
More, More, More
If you like this then you might like my other substack, Surviving Corporate, which is for people who are struggling in a corporate career, or dealing with the aftermath and recovering from the damage. It’s based on my own experiences and the things I’ve learnt along the way.
This week’s post, Eaten by Lions, is a draft of the first chapter of my ‘Corporate Survival Guide’ book project. Further chapters will be published every other week, alternating with posts about my experiences, learnings and thoughts of surviving corporate (well, actually not surviving and then recovering).