Daniel's blog

Our thoughts can be either complex or complicated or both. One can speak in a really clear way or in a very complicated way. In either mode, one can be saying very simple things or very complex things. It is therefore important to distinguish between  these two adjectives. This blog post is on speaking about complex things in simple terms and how we often get it wrong.

If we know only a little about some topic, we speak about it in simple terms if at all, using our everyday language. This is because we do not know what we do not know, so we by default have to keep it simple. It is worth noting however, that simplicity does not always equal to brevity for some people can take a really long time to convey a simple thought they do not even know much about.

When we dig deeper into a topic and we learn more about it, we start understanding parts of it. But we also realize that there is so much more to learn about it – we know what we do not know. This is the most tricky phase. Because we are eager to use all our new knowledge, we are no longer able to speak in simple terms only. Additionally, by trying to acknowledge and compensate for the known unknowns, we start speaking in a really complicated way. Our message becomes a lot less comprehensible.

It is only when we become really familiar with the topic that we are able to speak about it with clarity again. We know what we do know. We are able to pick only aspects most relevant to our audience and convey a complex thought in simple terms. This is the ideal phase. I always enjoy coming up to somebody with deep knowledge of an unusual field and asking that person to tell me in couple sentences about what they do. If they really understand their stuff, they are usually able to do so.

It is important to be aware which phase we are in. If we are in the first one, we should be careful not to be making any big claims about our knowledge of the topic. We should acknowledge we do not know much about it, speak briefly about it, and ask questions or learn more about it. If we are in the second phase, imagine how lost the poor guy on the other side of the table must be. We are now in a position to synthesise and to ask better, more targeted questions about the topic so we should do that rather than start rambling for ages. Finally in the third phase, we can share with others our gift of understanding complex topics in beautiful simple terms. I think this makes much better impression than using complicated terms just to show off we know them.

And always remember what Albert Einstein said:

“If you can’t explain it to a six year old, you don’t understand it yourself.”

Randomness is one of the key concepts the nature rests on. The very fact you exist, the fact you are reading this blog post or the fact you (might) know me is random in essence. Some people call it fate however. I have been hearing this word a lot lately. It happened because it was meant to happen. Or, if it is meant to happen, it will happen. While I sometimes very much enjoy this romanticised perspective on the world as well, I have always been struggling to fully buy into it. Things do not just happen because they are meant to happen; there must be something more to it. This blog tries to uncover this topic using concepts of randomness, exposure and ability to spot and to capitalise on opportunities.

Imagine a hypothetical example of you meeting an old friend you have not seen for ages at a party you were not intending to attend and he ends up getting you a job in his friend’s company. You take it, you really like it, your new boss loves you and in fifteen years you assume his/her job. At that point, your grandmother might say something like ‘you were always meant to become the boss of this pharmaceutical company, it was your fate and I am so happy you are doing it now!’ Was it really? Let’s look at this example more closely.

Firstly we have to accept that lot of things in the world happen on a random basis. The fact that your friend you have not met for years was at that party was random as was the fact he knew a person who was looking to hire somebody like you at a time when you were job hunting. So try getting comfortable with the fact that life throws a lot of stuff in your direction and that it does so on a very random basis. This is the bit we cannot control.

Where fate originates

The other two elements – exposure and ability to spot and to capitalise on opportunities we however can control. Since things happen on a random basis, some bad and some good, the amount of good opportunities you came across becomes a game of numbers. The bigger exposure you have, the more opportunities come your way, a portion of them being good. By eventually coming to the party as opposed to staying at home, you increased your exposure.

Finally you can control also the third element which I call ability to spot and to capitalise on opportunities. Because even if you had a massive exposure and were therefore getting lot of opportunities but you were not able to distinguish good ones from the bad ones and to pro-actively pursue the good ones, you would still not get anywhere. Once your old friend mentioned to you he knows somebody who is hiring, you called him up the next day, took a number on that guy, got in touch with him and persuaded him to hire you.

This is what truly makes a difference and is at the root of why you are where you are. Calling your life fate is suffering from a retrospective illusion of predetermination even though what actually happened was just a combination of random events and (lack of) your conscious effort.

The main takeaway therefore is that while we cannot completely control what random opportunities come our way, we can actively control how much randomness we expose ourselves to and where we do so. Then it is up to us to recognize the good opportunities passing by and to grab them. This is where I believe fate originates.

Note: If you have a mathematical mind, you can imagine fate as a product of a function with a random element in it and two variables. You can control the result you get by increasing/decreasing those two variables.

The word ‘rationality’ often conveys sense of maturity, responsibility and diligence. Gut feeling, heart or instinct are on the other hand associated with being daring, unreasonable or outright foolish. I would like to argue in this blog post that rationality is a very elusive concept which has many limitations and should not be applied to all problems. Relying on rationality in decision making can at times lead only to formalization of our instinctive errors and an illusion of control which is more dangerous than accepting from the start that the decision we are making is based on our gut feeling and treating it therefore as such; that is with extra alertness.

Limitations of rationality

Rational decision making is basically designing a mental model of what can happen in the future on the basis of which we subsequently make a decision. There are however in my opinion three main flaws of rationality. Firstly it is bounded by our own experiences, secondly we can never fully de-bias ourselves when estimating probabilities of future states of the world and possible consequences (most decision problems are unique and we therefore lack any objective reference point) and thirdly we are limited by our own cognitive abilities to process complex problems.

As for the first limitation, rationality is making the best decision under a certain set of assumptions. And here lies the problem – our assumptions are inherently imperfect because we can never know everything. We are basing them only on what we know yet there is so much more what we do not know and what we are not including in our model. Since everybody is unique, rationality is also individual and thus not universal. All the claim rational decision can therefore really mean is the most sensible thing to do given my very limited conscious knowledge of the problem.

The second problem is that rationality, despite its name, is hardly ever objective. Most decisions we face in life are one-off decisions (either because the decision itself is new or the context has changed). Since we have no objective reference points any probabilities we assign to the possible future scenarios in our model are therefore purely subjective. This means they are likely to be biased in just the same way as our instinctive decision making. Even though the fact we assess every probability individually can make us assess them better by devoting it more time, it is not sure this actually leads to better estimations. What it does lead to however is formalisation of our biases which can lead to wrong decisions looking as the rational ones.

Finally there are limitations given by the ability of our brain to consciously process large amounts of information. When facing messy problems we are likely to omit some important factors in our model which can again lead to a wrong outcome posing as a rational one.

So what should we do?

You might be asking at this point so what is the alternative. I believe that when facing complicated non linear and hard to define problems our brain works better in the sub-conscious mode than in the rational one. This is how animals think and how humans have been thinking for a long time as well.

It is only recently that we have started to formalise our thinking and while this has lead to great developments in our society, most of these developments happened either by chance or in an expert mode. That is by people who were experts in their fields and therefore had an in depth specific knowledge which they could utilise for rational decision making – they had a good set of assumptions. Most decisions are however not restricted only to one or two fields in which we posses expertise and so there is no point in pretending we consciously understand the problem.

We should therefore accept that our assumptions are imperfect, that we are biased when building the model and that we are unable to fully consider even the limited set of assumptions we have and not take the outcome of our rational decision making as the only possible solution. Letting our brain process all the information itself and trusting our gut feeling might be just as effective. I would like to conclude with an excerpt from Steve Job’s commencement speech at Stanford University in 2005:

‘And most important, have the courage to follow your heart and intuition. They somehow already know what you truly want to become. Everything else is secondary.’

Note: There are many problem structuring techniques to help overcome some of these disadvantages, in fact it is what I am doing my masters at LSE in right now, but they are often impractical to use in real life becuase of their complexity hence this post!

One of the key skills of a facilitator or a manager working with a group is to make the people understand what they are doing and why they are doing it – in other words – to make them feel comfortable about the process even though it might not immediately appear like the group is moving towards the intended outcome. One can apply this to his/her life as well to better understand where he/she is right now. While we can group these processes along many dimensions, the one I have found most useful divides processes into divergent and convergent ones.

The whole idea is that first we go wide, we diverge into many options, and then we go narrow again, we converge towards a solution. To illustrate this on a simple example, we can look at a process of trying to achieve a certain aim. Such process usually goes through several stages. A very basic divide would be:

1. We have a rough idea of what we want to achieve
2. We come up with many options of how we can achieve it
3. We select the best ones by applying various filters to all the options we came up with before
4. We execute those options

If we were to draw what is happening, it would look something like this:

Divergent processes

Firstly we diverge – we are creative, we brainstorm and generate as many alternatives as possible. This is very important because even if we would have very good criteria for choosing the best alternatives afterwards, we can end up with a bad outcome if we did not have any good alternatives to start with.

The problem is however that some people do not feel comfortable with this part of the process. After some time, they can either feel like they are losing time because they think sufficient number of alternatives has already been generated and they want to move towards choosing the best one or they feel uneasy and stressed about the sheer number of options, thinking that it might be impossible to ever come up with a simple solution. This is where the facilitator needs to step in and make sure that everybody understands that we are exactly where we should be at, in the divergent phase, and that we will converge in order to come up with a solution a bit later on. This makes people more comfortable with the process and allows them to focus on the problem more and to ultimately come up with a better outcome. Which is where the convergent phase comes in.

Convergent processes

The task is to narrow down the number of option to only a few best ones  in the convergent phase. There are multiple ways of doing so such as applying certain filters (having cut-off points) or if they are mutually exclusive comparing trade-offs between the options .

It is worth keeping in mind that these processes can often be found within each other as well. For example when starting with only a basic idea of what we want to achieve, we might diverge in terms of our objectives only to converge to the most important ones later on and to subsequently start the whole process again with diverging into possible ways of achieving those.

I have used this concept many times during discussions with teams I lead to make sure we consider all relevant options and choose the best one. I have also been thinking about my life over the last couple years in this way. I have been travelling to many countries, trying various things and meeting lot of people in order to experience what all is out there in the world and to see what I really want to care about in the future.

So next time you feel a bit lost or feel like your chosen alternatives are not really that good, think about what part of the process you are in and decide if you should stay there a bit longer or if you should go one step forward or backward.

I have been coming across the word context a lot lately. So much so, that it really forced me to think about it and to appreciate how vital it is to our ability to make good decisions as well as to understand decisions of others and the world around us as such. Is £1000 a lot of money? When your friend was angry some time ago, was she being unreasonable? Is 15 degrees a warm weather?

Well, we cannot answer any of those questions without knowing their context. Extra £1000 is a lot to a student living off couple hundred pounds a month but not to a company for which if might represent only a 0.1% increase on profits of one million. Your friend might have been unreasonable if the cause of her anger was having to work overtime on a one off basis but not if her boss shouted on her for no apparent reason. 15 degrees is pretty cold in summer but unusually hot in winter. This might all seem obvious but I would bet we don’t always take the time to understand the context before making or perhaps even more importantly, interpreting one’s decision.

Making a decision…

When making a decision, we need to realise what are the options we have realistically have, what consequences would they have, what is the likelihood of them working out and how would they be perceived – we need to know the context of the problem. If we come up with a model solution but we cannot implement it because of a specific organizational constraint, it’s a useless one. In a similar fashion, if we do not understand how the final decision maker will perceive our recommendation and if we would not be able to secure her buy-in, even the ‘best’ solution in the world would be useless in that context.

Organizational models (see for example my previous post here) or personality models, heuristics and preferences (e.g. if the final decision maker is risk averse, if she prefers quantitative or qualitative analysis, what are her key concerns etc.) are usually good starting points for gaining deeper understanding of our context.

…& interpreting a decision

Being aware of the context becomes even more crucial when we are interpreting someone else’s decision. What is an overreaction in some situations is an appropriate response in other situations and vice versa. Big risks can go unchecked and relationships built over a long time can go sour if we wrongly evaluate somebody’s decision. This can be because we were too lazy to dig deeper and understand the context in which that decision was made before we start judging it or because we misunderstood what message the other person was really trying to put across.

So remember, although it might sound obvious, context matters and it often matters more than you think!

I have recently been thinking about to what extend we can influence what is happening to us, if we should do so and how we should react if things don’t go as planned. Here is what I came up with.

Let me start more generally. When making a decision, we either do what we feel is right or we think of our objectives, generate some alternatives, assess how likely they are to help us in achieving those objectives and decide accordingly. From there on, things can go either well or not so well. It is however very important to separate the decision making process and the decision itself from outcomes of that decision for we can directly influence the former but not the latter. It follows therefore, that if we took the best decision possible at the time when we had to make it, we should never regret having taken it regardless on the outcome of that decision.

Influence what you can and don’t worry about the rest

With this in mind we can see that there are certain things we can influence by the decisions we make and our actions resulting from those and certain things we cannot influence – they just happen. You can call it fate, good or bad luck or whatever is appropriate in your culture. The first important point here is that you have to accept that there are things you cannot influence and therefore you should not worry about them. You should just accept them. The trick then lies in where we draw the line.

And here comes the second point – always do the best you can. That is the the best we can with the resources available such as time, intellectual capacity, money or influence at the time. You will have to make some trade offs but if they appear to be the most sensible actions to take at the time when things are happening, then that’s exactly what they are. If things go well, perfect. If they don’t, we should learn from the past but never regret having done what we did. The dots might connect just a little bit later on.

Here is a recent example from my life. I got to a final round of interviews with BCG, a leading strategy consultancy, and I really wanted that job. I spent almost all my time in the three weeks preceding that interview preparing for it and when I was entering their offices, I felt I have prepared the best I could have. Eventually, I did not end up getting an offer from them. I could have been regretting the fact I did not get it or that I spent all that time in preparation for nothing but that would defeat the whole point. I did the best I could have done and what made most sense at the time. I had nothing to regret and was able to look forward to other opportunities. I now have two other great consulting offers and most importantly, a clear mind.

So to summarise, the key to never regretting is acting in a way so that we never have to. That is doing the best we can right now to maximise the sphere of our influence and accept that from there on, things will just happen and there is nothing we can do about it.

I have just read an interesting paper on three basic decision making models explained using an example of the Cuban Missiles Crisis. In October 1962 the USSR placed offensive nuclear weapons in Cuba prompting the US for a reaction. The paper looks at how the US arrived to a decision for naval blockade as opposed to e.g. an air attack. In doing so, the author describes three basic decision making models which I believe can be applied in any organization. Knowing them can help you as a decision maker to make better decisions or to influence others in making their decisions more effectively.

Model I: Rational Policy Model

This model assumes organizations have a unified set of preferences and they make choices that would maximise their benefits. If you want to influence decisions being made under this model, try changing the costs associated with certain alternatives. The higher the cost (and consequently the lower the net benefit) of executing a certain alternative, the less likely it is to be chosen.

Model II: Organizational Process Model

This model views organizations as entities with codified operational practises and standard patterns of behaviour. Decisions made are therefore not deliberate choices of leaders but rather outputs of pre-established systems. If you want to influence decisions being made under this model, you have to change organizational practises in advance.

Model III: Bureaucratic Politics Model

This model views outcomes as emerging from ‘perceptions, motivations, positions, power, and manoeuvres of principal players’ (p. 630) within an organization. If you want to influence decisions being made under this model, understand internal politics of the organization. Who are the key decision makers? What do they really care about/what are they key concerns/what are their real agendas? Who has biggest influence over them? How can you form a powerful internal coalition to support your idea?

None of the models is perfect. The first one for example assumes that agents are rational which we know they are not and that organizations have only one set of objectives. The second one assumes absence of active decision making when unexpected situations occur and relies only on standard situations. The third one assumes decision making is all about politics and not about maximizing real benefits for an organization (here it is perhaps closest to the reality).

These models do however provide three distinct perspectives on decision making processes. I think it is beneficial to apply all of them in turn to a problem at hand as interesting insights about how to solve it/influence it might emerge.

The full paper can be found here.

This post is about one of the best books I read in 2012 – The Power of Habit: Why We Do What We Do in Life and Business by Charles Duhigg.

Habits are present everywhere – in our private lives, in businesses as well as in whole societies. They are an integral part of our everyday life and often supersede our rational decision making. This is beneficial because it shortens the time we need to make decisions (such as how to tie a shoe lace or the sequence of actions when we brush our teeth) thus allowing our brain to dedicate its processing power to other activities but it can be also detrimental to us (bad habits such as smoking or eating too many sweet things). Habits develop over time and once ingrained, different part of our brain controls them than the part responsible for our conscious decision making. Once triggered, they just unfold and guide our behaviour automatically. So the question is how to make the best use of this double edged sword.

The author argues we can put our beneficial habits to use and change our negative ones and he proposes a following solution. Firstly, we need to understand what habits consist of in general. Secondly we need to understand how our habits function and finally how we can alter them most effectively.

The habit loop

Every habit consists of three parts, together forming a loop. There is a cue, a certain trigger which makes us execute our routine, the routine itself and a reward we get from executing the habit and which incentivizes us do it next time we are exposed to the cue again. The author argues that it is very difficult to change the cue and the reward but relatively easy to change the routine if we keep the former two unchanged and gives a following example.

He would eat a chocolate cookie every day which he wants to stop doing as he is putting on weight. His routine is to stand up from his desk, go to a cafeteria to buy the cookie, chat with friends for a while and then get back to his work. So this is his routine; the next step to identify the reward.

Here he suggests we should experiment and try doing something else than our usual routine whenever we feel like executing it and then ask ourselves fifteen minutes later if we still feel the urge to do it or not. He tried going for a walk to a park, eating an apple or chatting with colleagues sitting nearby and then seeing if he still feels like eating a cookie. He realised is that what he really wants is not a cookie but a break to socialise with friends.

In order to change our habits completely however, we need to be aware of our cues as well. There are five main categories: location, time, emotional state, other people, and immediately preceding action. In order to identify the cue, answer the following questions whenever you do your habit. Where am I? What time is it? What is my emotional state? Who else is around? What action preceded the urge? After couple days, you should see a clear pattern emerging.

So the final step is consciously executing your new routine whenever the cue you identified happens and after some time the new habit will override the old one. After making himself go talk to a colleague for ten minutes at 3:30pm every day for couple weeks, the author stopped eating cookies at work all together.

How kebabs can trigger riots

Alternatively, you can try getting rid of a habit by inhibiting occurrence of the cue. When the US Army was still in Iraq, they had to intervene in several riots every day and for a long time could not figure out how to prevent them. Then one officer noticed by analyzing video tapes  that most riots are preceded by a small small crowd gathering on a plaza, gradually growing in size. At that point food vendors would show up to cater to the crowd, attracting in turn even more people. It is then enough if somebody throws a rock and a riot would begin.

Reaction of the officer was to ask the local mayor to keep food vendors out of plazas. The next time a small crowd gathered and started growing in size, signs of a riot were starting to emerge again. The only difference this time was that there were no food vendors and so after couple hours the people got hungry, dispirited and went home. The occurrence of riots decreased substantially.

Can you think of any ways you can change your habits or habits of your workplace/community for the better by a simple cost-effective intervention?

Have you ever spent hours or even days trying to solve a problem even though the solution was eventually simple and seemingly at hand? It has just happened to me. It’s two in a morning and I am writing a report for my statistical course at LSE. I am doing a regression model to predict birth weight of babies based on several factors. I did a few models but could not decide which one is the best one and have been stuck with it for the last hour and half.

At that point my girlfriend messaged me that she had a bad dream and cannot sleep. So we talked about it for some time and then she asked me how is my work going. When I told her I am stuck, she wanted to help so I had to explain the problem to her. And here goes the first lesson. Force yourself to be able to explain your problem simply and succinctly to somebody, who knows nothing about it. Try putting it in a text message! Mine was:

It’s very technical but basically there is a model that works very well only for some cases and so so for other and then there is one that works reasonably well for all cases. Which one to pick…

It forced me to strip the problem off all the technicalities and get to the point.

Then she asked if I cannot pick two models? A very simple question but a very powerful one at the same time. I’ve spent all my time before trying to improve or compare the two models mathematically but have not considered this one. And it is what I am going to do. For certain cases I am going to suggest the first model and for all other cases the second one.

The takeaway? If you are stuck, write down your problem in only few words and in a way so that anybody would understand what it is about. Then ask yourself some extremely simple questions about how you can solve it and give them proper thought. Great insights might emerge!

Another interesting person who recently spoke at the LSE was Stelios, the founder of Easyjet (and an LSE graduate). Here are some of the notes I took during his speech.

They revolve, albeit perhaps unintentionally, around the concept of Antifragility as developed by Prof. Thaleb and as outlined in my previous post. The world we live in is inherently volatile and unpredictable and therefore we need to structure our businesses in a way which would allow them to withstand this volatility.

Stelios does not believe in research, he prefers to learn by trial and error. Invest a (relatively) small sum, see how it goes and learn from it. Closely related to this is his second piece of advice – do not bet your farm. Learning by doing is all good but do not make a bet you cannot afford to loose.

Antifragility pertains also to his choice of industries to get into – the best markets are those benefiting from volatility. If the economy slows down, people start trading down and will fly Easyjet as opposed to legacy carries. If the market goes up, more people will be able to afford to travel and Easyjet will benefit as well. He was in a shipping business before which is very commoditised and when the market went down, everybody suffered.

The founder of Living Spaces, a room letting agency in London, spoke recently about a similar idea. It is best to focus either on the bottom segment of the market because if your product is relevant, the worst thing that can happen is that people will trade down, or on the other hand on the top segment where margins are fatter and better insulated from day to day volatility.

Finally, Stelios said that hard work breeds luck, a bit of cliche but a true one and that a good age to start a business is 28. By that time, one would have completed his/her studies, have worked for a couple of years and therefore would have identified what he/she is passionate about and are his/her strengths to build on.