21 concepts that have helped me as an Agile Coach
My job is to help organisations better respond to change and collaborate more effectively and sustainably to deliver value. Over the course of my career, some knowledge I have gleaned has been pivotal in helping me better understand how organisations and people work. This is by no means an exhaustive list, or a “how-to” guide, but hopefully captures some “pearls of wisdom” that might also help you identify and act on certain “smells” in your organisation.
⚠️ While I’ve tried my best to condense quite large topics, this is a longish article with quite a bit of info. I know medium readers tend to prefer shorter articles, but you can consider this a ‘reference’ post, one that perhaps you can save and come back to again and again.
1. Unorder and Order
This is a huge topic, that touches on multiple universes of knowledge including complexity theory, chaos theory, determinism, reductionism to name a few. At the risk of oversimplifying, I will attempt to summarise what I believe underpins most of the following concepts and observations.
We can broadly distinguish between two types of situation, the unordered and ordered. The former often deals with complex adaptive systems and chaos where there is usually large amounts of uncertainty present, outcomes are typically emergent and unpredictable with cause-and-effect only clear in retrospect or perhaps not at all. Ordered situations are considered simple or complicated and are linear and predictable in nature— when cause-and-effect can be known upfront.
The issue is when we mix these up (which we do all the time) and apply the wrong approach to the wrong situation e.g. when we believe we are facing a simple situation, but actually it is complex (or vice versa). As humans, we have a tendency to approach most situations as if they are ordered. It’s the easiest for us to understand and satisfies our innate desires for control and predictability. Unfortunately much of our world is complex and a lot of waste and damage is done when we fail to recognise it or attempt to force it to fit our ordered narrative. Dave Snowden’s Cynefin Framework is well worth learning.
2. Complex Systems
Extending on to the above, complex systems in particular, are not the sum of their parts. They are made up of a large number of diverse, interdependent parts that can adapt and change. A broken down car can be fixed when the faulty part is identified and replaced, but the poor performance of an economy cannot necessarily be traced back to any one particular part of the economic system. Instead, the parts of a complex system interact with each other in ways that cause emergent behaviours. These behaviours cannot necessarily be predicted or controlled because they cannot be isolated to any particular part of the system — they are a “feature” of the whole. This is why global optimisation is favoured over local optimisation and why complex systems tend to contain large amounts of uncertainty.
3. Individuals and interactions over processes and tools
It’s the first agile value, and may sound a bit cliché by now, but apparently it keeps needing to be said. People and collaboration are at the heart of any successful organisation and as social systems they fall firmly under complex systems and “unorder”, yet so few seem to understand how collaboration works and what makes people and teams tick (or conversely what process is and how and when to use it). Collaboration and creative thinking is often substituted by process because its a linear, ordered approach and creates the illusion of control. The problem is it very often tends to sweep under the carpet the humanity of operating in a complex social system, warts and all. It’s also no coincidence that most of these 21 concepts are related to people and their interactions.
4. Goodhart’s law
Goodhart’s law simply states that when a measure becomes a target, it ceases to become a good measure. Tracking metrics that relate to people’s performance can cause unintended consequences because we ourselves and the work that we do is often complex. For example, you might think it’s a good idea to evaluate employee performance against high priority goals with the hope this will motivate people to contribute. However, it will also likely result in increased politics, a loss of collaboration and fewer people willing to do the “thankless” tasks that keep the lights on.
5. The law of diminishing returns
The law of diminishing returns states that if you increase input, output will only increase in so far as other constant factors will allow. It’s from economics, but the general idea is applicable to most complex systems (and is somewhat related to the law of the minimum explained below). This can be seen in the effects of product planning. While we often believe that we will be more successful if we just plan more next time, to a point, it will not result in more return because our knowledge remains constant. We simply cannot account for unknown unknowns or known unknowns until we actually start doing. Knowledge in this case is a limiting factor. Other examples include writing lengthy documentation or job descriptions, developing intensive hiring processes and adding more people to a software project (Brooks’ Law).
6. The Cone of Uncertainty
The cone of uncertainty reveals that uncertainty is greatest at the beginning of a project. The theory is that any estimate made at the start can be four times early or late than the initial prediction (because this is the time when we know the least), but that variability inevitable shrinks the further into the project you get and the more you learn (emergence). This is why we try to minimise planning, reduce scope and batch size in order to deliver quickly.
7. Motivational Theory
In 1960 Douglas McGregor proposed Theory X and Theory Y. Theory X states that people naturally don’t want to work and are unmotivated followers that need to be lead by carrot-and-stick (similar to Taylorism). Theory Y states that people want to take pride in their work and be challenged. Management trusts individuals/ teams to take ownership and do the work themselves. While it may not be so black and white in reality, these are important approaches to identify because they tend to foster “mercenary or missionary” cultures, with the former running counter to contemporary thought on business agility. Specifically, focusing on extrinsic motivators for knowledge work tends to dull innovation and creative thinking. Other related and interesting ideas are Maslow’s Hierarchy of Needs, Herzberg’s Two Factor Theory and Dan Pink’s book Drive and his TED talk “The puzzle of motivation.”
8. Anchoring
The anchoring bias is when people place too much importance on the first piece of information they learn. An example is providing a stakeholder with an estimate at the very beginning of a project. Regardless of whether you stress it is just an estimate, or whether delays are likely to occur, you have just anchored them and their expectation. Why is this important to know? It is very possible that this “estimate” will be used to schedule dependent work or communications, and as the cone of uncertainty outlines, the variability of this estimate could cause a cascade of issues should it be incorrect.
9. The (upside-down) Iron Triangle
In traditional project management scope is fixed (often defined upfront). Time and cost become flexible (i.e. increase) in order to accommodate inevitable setbacks. The question posed is usually something along the lines of “How long will this take you?” Given the cone of uncertainty it’s akin to asking how long is a piece of string.
Agility is supposed to flip this on its head. Time becomes fixed through short iterations or batch sizes, as is cost (nominally in the form of the team remaining stable with no one being added). Scope then becomes flexible (i.e. decreases) in order to deliver something in a shorter time frame. The question then becomes “What value can you deliver within x weeks?”
10. Less is more
The larger the piece of work the more inherent risk and the more work we do in parallel (maximising capacity) actually slows us down. These are hard concepts to grasp and feel somewhat counterintuitive in a society where high impact and hyper-productivity is encouraged. However there is plenty of research that supports how doing less actually helps us do more (see Default Mode Network below). It takes great skill to reduce scope and slice work into smaller batches, just as it takes great skill to better prioritise, say no and “stop starting and start finishing”. Doing less isn’t necessarily easy.
11. Narrative Fallacy
This is the belief that we can explain the past through cause-and-effect when we hear a story that supports our beliefs. A good example of this is that we want to believe that success comes from hard work alone, but often this just isn’t the case (everyone has “unfair advantages” be it a rich parent, or being in the right place at the right time). The truth of the matter is that in a complex environment (with many variables), it is not always possible to know whether our actions in particular caused something or not. This doesn’t mean we should then just give up, but it should make you view things like blame (which implies someone is exclusively responsible for an outcome) through a different lens. This fallacy can be linked to our human penchant for causal determinism and the Fundamental Attribution Error.
12. Regression toward the Mean
The concept goes that in any series with complex phenomena that are dependent on many variables, where chance is involved, extreme outcomes tend to be followed by more moderate ones. In other words, really bad performance tends to be followed by better performance and really great performance tend to be followed by more mediocre performance. Let’s say your team completely failed to deliver. As their manager, you sat them down, did a post-mortem, identified actions, gave them a pep-talk and the next project they absolutely nailed. Sorry to burst your bubble, but while you’d like to believe it was because of your superior leadership skills (narrative fallacy), it could just be regression towards the mean.
13. Taylorism (Scientific Management)
Frederick Taylor (1911) believed “the manager knows best”, that it was the job of management to determine the best way for the worker to do the job, to provide the proper tools and training and to provide incentives for good performance. While this caused substantial improvements in an industrial setting, it came to be perceived as “mechanical and demeaning”, for highly skilled knowledge work, reducing employees to cogs in a machine. This tends to lead to command-and-control type management. Unfortunately we still are heavily influenced by this line of thinking and perpetuate it by often promoting employees to management positions because of their skill as an individual contributor rather than their leadership skills (Peter Principle).
14. The Feedback Fallacy
Marcus Buckingham sparked interesting debate in an HBR article suggesting that how we approach feedback is all wrong.
- We tend to be influenced by Fundamental Attribution Error and the erroneous assumption that “my way is your way” (egocentric bias). This makes us unreliable raters of others (Idiosyncractic Rater Effect).
- We believe that in order to learn we should see ourselves as an empty vessel that needs filling, however neurologically it’s more effective to focus on strengths, to build on something already there, than a subjective weakness that needs to be “fixed”.
- We think performance can be universally defined (think competency matrices). However while we might be able to agree on core skills, we are all different, with different innate or learned abilities, working on different problems in different contexts. In knowledge work there is no one right way of doing things. Creativity and diversity rule.
This doesn’t mean we shouldn’t give critical feedback, but we shouldn’t be surprised that it tends to trigger our fight-or-flight response and that this actually inhibits learning. On top of this, you are much less likely to act on critical feedback unless it was given by someone you trust. If you’re going to give feedback, it might pay to use non-violent communication i.e. emphasis your feelings and needs, rather than projecting your assumptions, values and judgement on the receiver.
15. Law of the minimum
The law of the minimum, or Liebig’s Law states:
…growth is dictated not by total resources available, but by the scarcest resource (limiting factor).
In systems thinking, being able to identify the limiting factors avoids improving the wrong things (waste). This is similar to identifying bottlenecks to flow in lean. Growth in a system is always limited in a finite environment but in many cases we don’t recognise what our limits are (and therefore tend to overshoot, or exceed the limits). We often don’t know because we’re either not looking at the bigger picture due to bounded rationality, or because there are delays or distortions in, or absence of, feedback. Both happen all the time in large systems such as organisations. Not only will there always be limits of some kind, some limits can’t be improved (like non-renewable energy sources) so it’s also about understanding which limits we can live with and where we can learn to become more efficient. A chain is only as strong as its weakest link.
16. Dunbar’s Number
Robin Dunbar has suggested that 150 is the cognitive limit to the number of social connections a person can maintain. While the number is disputed, it raises an interesting idea — it’s difficult, perhaps impossible, for us to maintain large numbers of stable relationships. What this means is that in big organisations, cohesion and collaboration naturally becomes increasingly difficult from a social networking perspective. Pan-organisational trust becomes a big issue as clusters of strong ties naturally develop around departments, products or functions and the risk of silos escalates. While it’s hard, if not impossible, to stop this clustering it’s important to recognise and minimise these effects by keeping the various groups exposed and connected to one another in order to maximise the benefit of weak ties.
17. The hedonic treadmill
The idea that an individual’s level of happiness, after rising or falling in response to positive or negative life events, ultimately tends to move back toward where it was prior to these experiences.
In other words, if you get a raise, it won’t be long before you become dissatisfied and want another one. Similarly when something horrible happens to us, eventually we become used to it. So while leadership may sometimes feel they can never make everyone happy no matter what benefits and improvements they bring to the organisation, there will simultaneously be truly less-than-satisfactory parts of the org that many people aren’t complaining about because they’ve become acclimatised to them. This is why it often pays to listen to the silence, not just the noise.
18. The Default Mode Network or (“Do Mostly Nothing”)
Related to “less is more”, maximising our cognitive capacity (often via multi-tasking and context switching) for a sustained period of time leads to increases in burnout and mistakes. There is also evidence that it decreases empathy, innovation and even intelligence. Physiologically, we need regular breaks. Time when we literally do nothing, or at the very least only one thing. This is when our DMN is activated, where our thoughts and memories are organised and the time when we are most creative. It’s why you have so many good ideas when you’re in the shower or just about to fall asleep at night. We don’t just need “slack” time for contingency, somewhat counterintuitively, we need it to get the most from ourselves.
19. The Responsibility Process
Christopher Avery developed a model called the responsibility process which:
…shows how all people mentally process thoughts about avoiding or taking responsibility.
We move from denial to laying blame, justifying our actions, then a feeling of shame followed by quitting or finally feeling obliged to do what we should have from the start. Ego plays a big part in initiating this process as well as hampering progress through it. Everyone has an ego, but letting it get the better of you can be one of, if not the, biggest detractor to collaboration. Ego doesn’t just lead to denial and blame (killing psychological safety), it also impacts your ability to trust others and others to trust you, and your ability to learn and take on feedback (Actor-Observer Bias). Unfettered ego is the enemy.
20. Ladder of Inference
The Ladder of Inference is a model of the steps we use to make sense of situations in order to act.
To summarise, every action we take is influenced by the meaning or conclusion we have drawn from an observation. While we may all have the same observation, we may take different meaning from it based on our own world view that we have built since we were born. It’s these situations that often generate conflict between us as erroneous assumptions are often imposed that we share world views. To get on the same page, we need to have healthy conflict by shelving our ego and discussing where and how our mental models differ. Any collaborative endeavour requires this from individuals in order to build shared mental models together.
21. The Principle of Consent
It’s not uncommon in the workplace for people to either implement a solution (often before understanding the problem) and without having asked for much feedback in order to move quickly, or to try to search for consensus in order to generate buy-in. The former often lacks buy-in and the latter speed. The principle of consent is a decision-making technique that helps us get the best of both worlds. The approach requires that you invite dissent around objections (as opposed to opinion or preference) from a core group of impacted individuals or representatives and rely on an implicit contract of consent i.e. if you choose not to actively raise objections even after being invited to, then you are consenting to follow through on the proposal.
“Adopting the principle of consent shifts the aim of decision-making towards identifying a solution that’s good enough for now, and where there are no obvious worthwhile improvements that would justify spending more time.”