Mastering Complexity: Part 2: Complex vs Complicated
- Introduction
- Part 1: What Complexity Is
- Part 2: Complex vs Complicated
- Part 3: Harnessing Complexity
In Part 1, we saw that today’s global business environment is highly Complex, and:
- Introduced the idea that Complexity sits at one end of a nonlinear spectrum with Simple at the other and Complicated between them.
- Explored how Complexity involves many, diverse and interdependent parts or – “agents” (typically people) – interacting autonomously in relation to Complex challenges and other Complex systems, and therefore subject to both known and unknown factors.
- Described how Complexity leads to constant self-reinforcing change – including and especially when it comes to Value – emergent properties of self-organisation and innovation, and inherent unpredictability and risk.
Throughout, we explained how Complexity is reflected in organisations and then summed-up by looking at the Complex PESTLE challenges that organisations face in the modern world when trying to facilitate Value for their consumers.
With the latter, we particularly saw how advances in modern information technology have uniquely empowered and accelerated the elements that lead to Complexity and its effects, creating a “perfect storm” of exponential and unprecedented challenges.
In this piece, though, we will see how accepted “wisdom” instead persists in treating the world and its challenges as primarily Complicated – even if they may use the word Complexity – and that this explains the growing disparity between what organisations are faced with and their ability to deal with it.
To do this, we will look in more detail at:
- What makes something Complicated and how it differs from Complex.
- The Complicated responses to today’s Complexity and why people pursue them.
- Why such responses have been ineffective at best, and – at worst – catastrophic.
As we did last in Part 1, whilst mostly talking in general terms, we’ll relate back to organisations and business.
1. Complicated vs Complex
We previously introduced the idea of “…a nonlinear spectrum, rich in meaning, that runs from simple at one end, through Complicated, to Complex at the other”.
We explained how this spectrum is applicable to systems, to challenges, and responses to challenges, and then defined in detail what we mean by Complexity.
To now illustrate the differences between the “points” on this spectrum – and especially why it is nonlinear – we will describe each in terms of:
- What defines it.
- What it is made up of.
- What it is like.
- What it results in.
Simple | Complicated | Complex | |
An example “system” | A car wheel | An engine | A driver |
An example challenge | Changing a tyre | Finding a fault | Driving on ice |
What defines it | |||
Its conceptual scope is… | …minimal / objective: | …wider / objective: | …complete / subjective: |
Its boundaries are… | …limited and clear | …broad and clear | …expansive and fuzzy |
Its surrounding context is… | …wholly static | …mostly static | …in constant flux |
Known knowns? | Yes | Yes | Yes |
Known unknowns? | No | Yes | Yes |
Unknown unknowns? | No | No | Yes |
What it is made up of | |||
Its parts / agents are… | …few / fixed | …numerous / fixed | …numerous / changing |
Simple parts / agents? | Yes | Yes | Yes |
Complicated parts / agents? | No | Yes | Yes |
Complex parts / agents? | No | No | Yes |
Parts / agents are diverse… | …sometimes | …often | …always |
Their autonomy is… | ….non-existent | …limited | …high |
Their interactions are… | …few / linear | …numerous / linear | …myriad / networked |
What it is like | |||
Value is… | …fully predictable | …fully understood | …emergent and fluid |
Change and risk are… | …non-existent | …possible / controllable | …inevitable / exponential |
Cause and effect are… | …directly connected | …eventually connectable | …obscured |
Predictability is… | …absolute | …high | …minimal |
Repeatable processes are… | …essential | …critical | …of limited use |
Expertise… | …is optional | …is essential | …may be helpful |
Effort… | …isn’t necessary | …guarantees results | …may make no difference |
Issues with it need to be… | …resolved by known answers | …solved by finding answers | …dissolved by change, but never “solved” |
Its results are… | …identical each time | …similar each time | …unique each time |
What this shows is that the principal difference between Simple and Complicated is one of scale.
Something that is Complicated may be very difficult, involve a lot of work, and have many parts or “agents” involved – with many interactions between them – and even appear to be Complex, but, ultimately, as with the Simple:
- Its surrounding context and circumstances are (near) static.
- It can be bounded, analysed, understood and precisely described.
- Interactions between its parts and/or “agents” are structured and linear.
- It operates to cause-and-effect rules that can be discovered and controlled.
- Specific aspects can be isolated and dealt with independently of the others.
- It can be readily and repeatedly observed, with success objectively defined.
Putting all this together: the scale is all that really varies between Simple and Complicated, but what happens with both is repeatable and predictable, and issues can be understood, broken down into smaller parts where necessary, and solved.
In contrast, though, whilst Complicated and Complex are usually used as synonyms (except in systems literature) – and whilst, as illustrated above, they share some apparent similarities and the Complex can have the illusion of just being “even more Complicated” – the differences between them are seismic and multi-faceted, i.e. this is where:
- We go beyond the known / knowable to also encompass the unknown / unknowable.
- We move from the linear, hierarchical and bounded to the networked, interdependent and fluid.
- Subjectivity is introduced and a threshold of autonomy crossed.
- Change is no longer proportional but exponential.
- Repeatability is mostly eliminated.
- The relationship between effort and result is severed.
Putting all this together: this is no longer about scale; the Complex is profoundly different in nature from the Complicated – in particular due to emergence.
Control and prediction are no longer difficult, but largely impossible, and issues (or “wicked problems”) not only resist attempts to break them into understandable parts but also ultimately being “solved” at all.
2. Complicated Responses and Why They’re Pursued
We have just seen that what is true of the Complicated is fundamentally not true of the Complex – especially when it comes to understanding and responding to challenges.
We have previously seen that Complexity is nothing new – it’s the nature of nature (and human nature) and has been around since the dawn of time – and also that today’s business challenges are increasingly Complex.
However, even if the word Complexity is increasingly used to describe today’s business challenges – e.g. in the C of “VUCA” – it is typically done so very loosely, without the precision and richness we are proposing (and, as noted in Part 1, without understanding the primacy of Complexity).
And the responses being made to challenges – assuming a response is made, as opposed to a retreat into dogma and denial – remain predominantly Complicated, i.e.:
- Treating their context as static and knowable, especially when it comes to what constitutes value, which is seen as something to be “created”, “added” or “delivered”.
- Duly focusing on structure, and prescribing pre-defined and repeatable solutions – organograms, traditional contracts, regulations, processes and standards.
- Assuming they can be understood and solved: whether through applying these familiar solutions and related tactics more rigorously or widely – especially risk management, surveys, KPIs and “balanced scorecards” – or the presumed panacea of new technology, e.g. the “Internet of Things”, AI and Big Data, the starting point is that analysis will fully explain them and yield actionable answers.
- On the reductionist assumption that the overall situation is simply the sum of its parts, attempting to use this analysis to break it down into smaller parts, and then doing one of two things:
- Addressing each separately (and often sequentially), e.g. with task-focused “red teams”.
- Trying to address most – or even all – of them at once on the assumption that dealing with the parts will inevitably lead to improvement and resolution of the whole, e.g. capability and behavioural development.
- Assuming that efficiency is the same as, and leads to, effectiveness, such as LEAN, Six Sigma, process re-engineering, etc.
- Relying on linear and hierarchical “command and control” structures and approaches to set strategy, cascade it “down” and force outcomes, such as refining governance structures, top-down commitments to change and/or core values, reducing spending discretion and enforcing contracts.
Why the Discrepancy Between Reality and Prevailing Wisdom?
To begin with, human beings – especially in the West, influenced by Plato – fundamentally generally don’t “like” Complexity:
- We naturally gravitate away from it.
- We prefer order to “disorder” (and, as F.A. Hayek observed, our brains work by creating rules and structures).
- We hold up perfected and rigid ideals and aim for them.
- We want things to be predictable and controlled,
- We like (and take pride in) understanding and finding solutions to challenges.
We are naturally predisposed to Complicated thinking, even if this over-simplifies reality.
This basic human trait was then accentuated by the scientific method – applied to business by the principles of Frederick Taylor’s Scientific Management – and by the Industrial Revolution, where scientific advances led the world to be understood principally as a “machine”.
Everything was understood to be governed by known, fixed laws – the understanding of which seemed to enable nature to be tamed, accurate prediction and total control – and people became largely non-mechanical and non-autonomous “cogs” in predominantly manufacturing processes.
This naturally led to efficiency being the principal threshold of competition, “exploitation” through repeatability (ahead of exploration), and optimisation and control being the keys to competitive advantage.
Whilst science has since moved on to see Complexity as even more fundamental than laws – e.g. in quantum theory or in understanding biological systems, both of which leading thinkers have related to business – the business world in particular has instead steadfastly clung to complicated approaches and their underpinning assumptions.
(Even to this day, the metaphors and similes used in connection with the economy remain mostly mechanistic: cranking “levers”, the “engine” of growth, etc).
This can be partly explained by the fact that the business world primarily involves People, and – as we have said – people naturally prefer Complicated thinking, even when it can’t address the unknowns that occupy their greatest challenges and fears.
It becomes even more understandable when remembering that, before the seismic impact of automation – and, even more so – that of modern information technology) the pace of change was a lot slower, which masked the effects of Complexity.
Paradoxically in a sense, technological advance has – in some ways – perhaps reinforced old attitudes.
As much as it has enhanced connectivity and interactivity, the “human” element has been reduced, which implicitly drives towards mechanisation.
Similarly, the instantaneous nature of interaction, powered by breakthroughs in connectivity and device ubiquity, collapses the elapsed time between now and the future, creating an illusion of predictability and control.
Also, even now, not all challenges are Complex, and for Complicated challenges (even if these are being increasingly and progressively marginalised and simplified by the persistence of learning):
- Whilst there is a sense in which the world has always been Complex – especially as People are involved – the “worst” that happens with Complex humans in a Complicated challenge is that they introduce error and inefficiency…
- …where Complicated thinking and approaches are absolutely appropriate and successful – especially where the primary focus is physical “products” and the “tangible”, which has historically been the case (culminating in LEAN production).
Complicated thinking is what has been familiar and accepted as “best practice”.
Reinforcing our natural preference, and “indoctrinating” it ever deeper, is the entire education system (including management training), there has been heavy investment in it – time, energy, political capital and reputations – and an established and persuasive consulting industry is built upon it.
Any “challenge” to the establishment is either easily dismissed or resisted – including as there have been so many management “fads” that have briefly burned brightly but were ultimately proved to offer nothing sustainable.
Related, Complexity is difficult and uncomfortable, creating new and unfamiliar challenges, and the need for agility at ever more fundamental levels. It is also is a relatively new field – even now, much literature is analytical or anecdotal and not practical – so there hasn’t been an alternative approach to date.
The assumption is then (somewhat understandably) made that, without total control, there is chaos and anarchy.
Taken together with the fact that change is always difficult, reinforced by the organisational structures that have built up – and the fact that traditional approaches sometimes appear to work for a time (even if silently making things worse for later, through hiding the real issues, accelerating denial and further entrenching the existing order) – all these factors have meant that there hasn’t previously been any strong or clear driver for change.
They have also fed the assumption that – where problems were encountered – the fault wasn’t in the approach but rather in its application, and so it is applied even more forcefully and often in a “defensive” way, which (coupled with vested interests) also then makes change even harder.
The problem is that this is no longer tenable: the gap between today’s challenges and the ability to deal with them – sometimes called the “productivity puzzle” – is ever widening. But why?
3. Why This Is Ineffective, Even Catastrophic
At the most basic level, Complicated approaches to Complex challenges fail because they are based on a set of fundamental and related (mis-)assumptions about those challenges – insidiously nurtured by the “scientific” method – that proves to be fatal, i.e. that:
- Challenges are knowable and to be anticipated and “solved” (or, with contract and risk management, mitigated against)… when, in a time of accelerated change, it is increasingly impossible to look ahead; instead, resilience and adaptation are most important.
- They can be clearly defined, are mostly tangible and remain static… when Complex challenges and value are always amorphous, dominated by intangibles, and in a state of rapid flux, such that interventions often have unforeseen – even disastrous – consequences, and organisations need to prize agility.
- They are the sum of their parts… when the multi-dimensional relationships between the different aspects of Complex challenges, and their associated property of “emergence”, means that such challenges are always more than the sum of their parts.
- Effectiveness requires prioritising efficiency and optimisation… when constant change makes this impossible, and Complex challenges reach far beyond the operational, i.e. vs .
Indeed, not only are “efficiency” and “effectiveness” not the same thing – as Peter Drucker said “efficiency is doing things right; effectiveness is doing the right thing” – but Complexity decouples them to such an extent that focusing on efficiency is not just sub-optimal but counterproductive.
Far beyond the truism that doing the wrong thing “well” is often worse than doing nothing, efficiency by definition optimises what is known and already in place – i.e. it is developmental, focused on the objective, the comfortingly familiar and the already known (sometimes referred to as the Streetlight Effect).
It also ignores that challenges and changes are affecting the “higher” levels more often and more rapidly, where disruption is far more profound, and where these levels were previously relatively stable and slow-moving.
A focus on efficiency is therefore severely limiting in a climate where change is constant, dominated by the fuzzy and subjective – and what isn’t already known, both with means and ends.
The unprecedented impact of technology has let the Complexity genie out of the command-and-control bottle, the mismatch between what is valuable and what is measured and focused-on is getting wider and wider, and Complexity can no longer be ignored or set aside.
The threshold of competition, competitiveness and competitive advantage has moved.
In organisational terms, typically not only are the approaches taken Complicated, but so are the structures adopted – as an inevitable consequence of reductionism – with levels of formal “command and control” hierarchy through which strategy, decisions and information flow.
Considering that efficiency is often the goal, a bitter irony here is that such structures are extremely inefficient – creating a fixed and extremely tight bottleneck at the “top” through which everything must flow “down”, and delaying and distorting flows back “up” from the ground.
Such structures also directly work against the agility that is essential for effectiveness – agility requires change, which requires innovation, and we saw in Part 1 that innovation (or “emergence”) results from a high degree of information-rich networked connectivity between a numerous and diverse population.
With formal hierarchical structures, the “population” is artificially constrained, connections are hierarchical and few, and information is limited.
Further reducing effectiveness is the implicit assumption that these few at the top are best placed to understand, use and share information and act, which – often through no fault of their own – is rarely the case.
To begin with, unless actively compensated-for, decision-making is heavily skewed by our mental models – not only have we seen that these mental models gravitate to the overly-simplistic and that they are significantly dictated by prevailing wisdom as to what constitutes “best practice”, but they are slow to change in response to new information, which is increasingly problematic when change is so rapid.
This is a challenge for anyone involved in decision-making, but it is especially acute when most of the people involved in working with the challenges being faced are “disenfranchised”.
At best, their input is diluted and out of date by the time it filters back to the top; at worst, they simply aren’t consulted; either way, antipathy is created.
From here, a sense of responsibility is diminished: it is “someone else’s decision” that is being enacted or “someone else’s process” (or “the system’s process”) being followed, and the need to think and act independently in response to circumstances is proscribed or removed.
Whilst Complexity means that each situation is significantly different, repeating the same ways of doing things reduces learning and heightens the risk of complacency, even when those things demonstrably aren’t working (particular aspects of the Grenfell fire in London, Boeing 737 crashes and responses to the London Bridge terror attacks being tragic examples within recent memory).
In part, all of this reflects that formal hierarchical structures, bureaucracies and process-centric approaches introduce further distortions and unhelpful behaviours when it comes to delivering value.
A key concern for a formal hierarchy is preserving itself and its investments and, within it, a key concern inevitably becomes satisfying the level or levels “above”, which may or may not be the same as what’s good for the organisation in the situation and which haemorrhages focus.
All of these reductions in effectiveness caused by responding to Complex challenges with Complicated approaches are reflected in this modelling of their typical implementations:
Projects and initiatives run late and over budget, relationships are under strain and most organisations are all too familiar with any or all of the following:
- Management and leadership challenges: unpredictability, missed opportunities and early warning signals, overwhelming amounts of data, difficulty of setting goals and making decisions, constant pressures to do more with less.
- Governance and quality issues with poor execution and performance, inconsistency, difficulty converting strategy into best practice and in articulating and communicating what good looks like.
- Front line issues with dissatisfaction with leadership decisions, repeating problems, mismatches between what strategy and leadership prioritise and what actually works in practice, a sense of disenfranchisement and of being ignored, and a frustration with temporary fixes and management “fads”.
As we’ve seen, these are all inevitable and understandable – especially with hindsight – but not for those anchored in familiar mental models and working practices.
Indeed, in the face of these difficulties, most see the resulting mediocrity, waste, difficulty and failure as simply unavoidable – an inevitable “cost of business” – whilst some place their hope in the seductive hope of technology and data.
Whether or not this is because they recognise that technology and data have been part of creating the challenges (so ought to be part of the solutions), the assumption is that more of both (especially in the shapes of AI and Big Data) will somehow solve everything, even if:
- It isn’t clear exactly how (and despite the track record of technology never providing a panacea).
- Leading experts in the AI field have expressed doubts about its ultimate potential and question whether the hype is fully justified.
- Awareness seems to be slowly growing of where AI and Big Data are useful and where they’re not, such that the panacea balloon is showing some early signs of deflating.
- Technology is, of course, built by humans and susceptible to their biases in how it works.
- Everyone has access to it, nullifying potential competitive advantage.
- Data can provide inputs, but doesn’t ever actually make decisions itself.
As one report put it, “Viewing technology deployment as a quick fix to long-standing neglect can put a project on a path to nowhere”.
Indeed, there are plenty of justifiable claims for how technology can automate, help scalability and increase efficiency, but there is no specific or sustainable claim yet made – beyond a general one that efficiency frees up time – for how technology can help increase effectiveness within Complexity.
The results of ill-thought-through investments here have been unsurprisingly disappointing.
Ultimately, though, this continues to reflect a Complicated mindset, where the challenge is one of analysis to find a solution, with the responsibility for doing so significantly “abdicated” to a “system” (and hence its seductive appeal) – failing to respond to the facts that:
- Complex challenges involve human beings, such that human intelligence is needed (including to try and forecast, where Big Data simply isn’t helpful).
- It is the quality and focus of data that matters far more than its volume, or the ability of software to process and derive things from it: new, People-centred data is needed.
At this point, it is worth acknowledging that People are sometimes made the focus – especially the current preponderance of “behavioural training”, vague emphases on the importance of “trust”, the emergence of “relational contracting” in business relationships, and the increased time spent in workshops.
This may well reflect an almost-certainly implicit acknowledgement of one or both (i) the limitations of (un-)Scientific Management, and (ii) the role of people (and the connections between them) in creating and responding to the challenges faced.
However, Complicated thinking ultimately dominates because either or both of the following is true:
- The focus remains “traditional”, focused on Complicated goals of problem-solving or behavioural development – i.e. “capability”, “behaviours” or “trust” become catch-all terms, people just need more training in the right methodology and all will be well (which, as an important aside, implicitly “blames” those involved: not just “insulting”, but refuted by an understanding of complexity).
- It is typically a “bolt-on” to traditional approaches with very limited (or no) persistence, resilience and ability to scale, e.g.:
- “Relational contracting” is a repackaging of mostly standard contracting practice.
- Focusing on behaviours, but without recognising that they are significantly products of other factors, which either remain unaddressed or addressed with traditional methods.
- A use of apparently “consultative” tools – surveys, etc – when the “real” value is perceived to be “getting people in the room and talking” to find out what’s going on and “solve” things. Not only is “solving” a “complicated” way of seeing things, but – whilst there are certainly times when starting conversations and airing issues is needed (especially early on in a change process) – workshops like this ultimately can’t scale.
At best, then, whilst a focus on technology, on people, or both, may represent something of an instinctive realisation of their role in creating current Complexity – and thus their role in addressing it – these are only partial, flawed and often tokenistic responses.
As a variant of the quote often attributed to Einstein puts it, “We cannot solve our problems with the same thinking we used when we created them”, and a new approach is needed to respond appropriately to complexity (alongside continuing to meet complicated challenges with the many effective tools that already exist to do this).
Because Complexity is inherently transformative – through emergence – the response to it equally needs to be transformational. Or, to put it another way: a paradigm shift is needed.
But what does a Complexity-informed and Complexity-driven paradigm shift look like?
It involves a new mindset, a new toolset and – through applying both – the development of a new skillset.