Sun Tzu and the art of project management

I really like Sun Tzu's the Art of War and to understand it better I like to analyze various things through its lens. Here is something I originally wrote internally in our company newsletter about Sun Tzu and the art of process automation, with a special focus on RPA or Robotic Process Automation. But the advice is meant to be more widely applicable.


This week's tip of the week is a bit different. On the surface level it advocates that you should think twice before assaulting fortified enemy positions, but it comes with a long preamble and turns more contemporarily applicable before the end.


Sun Tzu's the Art of War dates to 5:th century BC and is one of the oldest books that is still widely studied and read. Before diving into the specific section I wish to talk about today, we need to set the scene a bit.

First of all, we should note that even though the title talks of War, it is not a mere military manual. In the cultural context where it was written, the way for a philosopher to get funding was to write about military strategy. So, anything deep you wanted to work on was better to be written in a format that generals would like to read. Thus, we get this classic book on the philosophy and practice of how to work in complex situations where different actors have different and possibly competing goals. It just so happens that this is very applicable also in the theatre of war, so here we go.

Secondly, the book has (at least) two levels of metaphor. The first level is in the Chinese language which, together with its writing system, is much more conceptual than most European languages. For example, the character 领 (lǐng) can mean both the noun "leader" or the verb "to lead" depending on the context. Furthermore, when combining this word with others we can create new concepts; when 领 is combined with 袖 (xiù), which means "sleeve", the resulting word 领袖 (lǐngxiù) refers to a figurehead or a leader with significant influence, suggesting someone who leads by pulling others along with them, as if by the sleeve. I leave it to you to extrapolate the amount of expressiveness these kinds of constructions might lend to a language. (I do not actually speak any Chinese, but I can only assume that they also have an incredible collection of puns.)

In the second, deeper, level of metaphor Chinese is what e.g. Erin Meyer (the author of the Culture Map) calls a high-context language. By this we mean that it is common to talk of many topics indirectly, so that only by knowing the context you can extract the full meaning. For example, in the Art of War by "terrain/ground" we mean both the physical ground where the literal troops might be moving, but also more generally the non-changing aspects of the situation, whereas by "sky/weather" we mean both the literal weather conditions but also the volatile changes in the environment that are beyond our control and hard to predict. Thus a scholarly colleague might say to you after a client meeting that "even though the weather was pleasant the muddy ground was hard to walk" and it would be your job to intuit that in this context they are saying that there has been no unexpected changes (weather is good), but the customer's overzealous commitment to use an old version of the invoicing system everywhere is causing delays in the project (the ground is hard to traverse). Or perhaps they just visited the customer on a sunny day and the muddy ground ruined their shoes – you need to know the context.

These two levels of metaphoracity bring about several important notions. One of them is that with these notions it is even easier to see why the book might be about more than just war. Another is that the book (and often other Chinese philosophical literature) can be exceedingly hard to translate to a less subtle language like English. For us an important notion is that though the Art of War is a short book, it packs a huge amount of information as pretty much any line or paragraph can be read with different interpretations. The detractors might now say that reading the Art of War is akin to fortune telling or tarot – with enough interpretability you can fit whatever you want to say to some cherry-picked passage of the book. I partially disagree and believe that even though you can misuse the text in this way, there are a lot of true insights encoded in the myriad of interpretations. Furthermore, even if it was the case that you can fit any message to this frame, then I claim that it will help you at least analyse the situation better if you view it through the lens of Sun Tzu's work.

Today's passage

Anyway. My favourite passage goes as follows:

The best policy is to attack the enemy's plans.
The next best is to disrupt alliances.
The next best is to attack the opposing army.
The worst is to attack the enemy's fortified cities.

This passage can be found from Chapter 3 just below the more quoted lines of "To win one hundred victories in one hundred battles is not the acme of skill. To subdue the enemy without fighting is the acme of skill." And to me my favourite passage feels like an instructional complement to this more famous passage – though again this will be just one reading of the situation. An important thing I want to emphasize before any deeper analysis is that just like the rest of the Art of War, this advice will be very applicable to situations outside confrontational ones. The opposition in this metaphor does not need to be a e.g. customer or your boss during salary negotiations – it might be that your opposition is an automation problem that you are 'fighting' together with the customer. As discussed in the introduction, the passage can have several different interpretations, but the way we look at this today will be that of "don't hit the thing that has been made especially resistant to hits." Let's look at a few examples.

Example 1

Suppose the challenge in the automation process that you are creating is a third-party MFA+Captcha system that the RPA struggles to authenticate with. The Captcha-protected MFA is the fortified city and it will be very hard to attack. But with the customer you can then try to think if there is another way to log in as a user; maybe there are some credential systems? This would be attacking the army. But perhaps a full login is not even required? Maybe the data you need to get from there can be extracted from a database with an API? And you should ask if, when, and where this data is even needed. This is all moving away from the fortified city and towards more generic targets of alliances and plans. So, thinking through the problem through this scope we see that we should always figure out what it really is that we wish to accomplish? Usually, our end goal is not to destroy a fortress for the sake of destroying a fortress – it is merely means to an end and we should ask if there are any other means available. Figure out what we really want to achieve with the customer, and keep asking why some goal is desired to figure out what the real plan is. Though I would also hazard to say that it can cause different problems if you spend too much time figuring out the optimal strategy instead of doing something – this can incur costs in e.g. lost initiative. There is a balance on figuring out the trade-off between acting without thought and being stuck at the drawing board. As the Zen of Python puts it:

Now is better than never.
Although never is often better than right now.

Example 2

Let's look at different kind of example – you are having a 'discussion' with a young child who does not want to take a nap. Here we can approach the situation as a more confrontational one as it is you vs the child. (Though it is you versus the child, your goal is the good of the child, interestingly enough.) Instead of trying to wrestle the child to bed and force them to sleep, a more avid reader of Sun Tzu might avoid that walled city and instead suggest that: "Okay honey, you don't have to go to sleep, just lie still in the bed with your eyes closed for five minutes and then you can go back to play." (Warfare is one thing. It is the philosophy of deception. – Sun Tzu.) You have avoided the battle over the fortified city of "change the kid's mind about taking a nap" and introduced a new setting where the kid gets to keep their decision by agreeing to a seeming compromise. They'll fall asleep not by your command but almost accidentally due to the circumstances without realizing that you were the one setting them. You have avoided the fortified city and lured the enemy army into a terrain of your choosing; always remember that walled cities are usually hard to move while fielded armies are mobile by nature.

From this example I want discuss a crucial point of the passage - it does not say that you should never attack walled cities, just that it should not be your first or even second choice. With a child, sometimes you will have to put your foot down to assert dominance – "No you cannot run outside without shoes!". And sometimes in hard negotiations you will want to be seen as the strongest player at the table who will enforce their will from time to time. But - you should always be aware of the costs. Crushing the one place where your opposition has declared they will not budge will make you look strong, but it can make your opposition lose face. Fear of this can produce unproductive hills that people are willing to die on. As Sun Tzu puts it:

Always leave an escape route for a surrounded army.
Do not press a desperate foe.
This is the art of war.

Another cost you must consider when attacking their walled cities is that by pushing where it is hardest to push, you might not only be missing an opportunity to attack their plan, but you might be actively working according to their plan. The opposition might be hoping or counting on you wasting your resources in assaulting the very visible but robust target while they have free hands to do something else. Think of a street magician who is asking you to make sure you select a card at random – you spend a lot of effort to make sure they are not affecting your selection, but you fail to realize that the deck only has one type of card and that the magician's assistant has already stolen your wallet.


My suggestion is that you read the paragraph again:

The best policy is to attack the enemy's plans.
The next best is to disrupt alliances.
The next best is to attack the opposing army.
The worst is to attack the enemy's fortified cities.

And in the future, you will come back and read it a third time. If you have the possibility, go to the library, and read it in a language that is not English.

Finally, note that Sun Tzu's book is not "The Law of War" but "The Art of War" – this is not a rule to follow blindly but a time-tested advice you can use to analyse your situation. It is important that you find your own point of view to the passage and internalize its core so that you can derive the meaning you need. Anyway, I want to finish by giving some more direct advice (an actual "tip of the week"). To this end I would suggest the following.

  1. Think hard about what the problem is you are 'attacking'. Why do you want to solve this problem? Where does it fit in the bigger picture? Are you fighting the army or a fortified city here? Cities are fixed in place, but armies can be made to move – can you choose your terrain more advantageously by surrendering the goal of the city but drawing out the army?
  2. Look at what kinds of 'alliances' the problem has. What factors in the project deepen or diminish the problem? Will someone lose face if this problem is or is not solved in a certain way?
  3. Try to see the big picture. What is the real plan here and what do we really want to accomplish? Remember that people don't buy drills, they buy holes in the wall.