Simplicity is underrated in the world of AI and ML solutions …and that needs to change.

Jeff Bezos in 2017 quoted:

“Innovation is not disruptive, consumer adoption is “.

What does this mean ? This means that the customer/end user does not give a sh**t about how weirdly insane or complex your product is. All they care about is whether it will solve their problem or not. What lies in the backend is not their problem. The reason they want to buy your solution is to simplify something at their end. If simplicity is so important for the end consumer, why businesses do not value it ?

A. Simplicity needs to be in the intuition, not necessarily in the product

Ya…I know the title above does not make sense so I will try to build some sense around it. In today’s world of AI and ML driven solutions, there are folks who write to me saying-“you keep saying that we need to simplify things but the underlying algorithms of AI and ML in themselves are complex and you HAVE TO understand that complexity to make the best use of these algorithms.”

That is true. But the simplicity that I talk about is different.

A1. The (self created 😁) theory of “Simplicistic intuition

My version of simplicity is not necessarily about a simple product but the simplicity behind the process of building that product.

How many time have you read a news in WSJ, read a case study, read a real world business news, were walking in a store, were waiting in line somewhere, and in the midst of all that English and jazz and chaos, you were able to see what exactly is happening, why is it happening, what will be state in x years, what are few things that need to happen to mitigate a business challenge, how does this impact a totally unrelated area- all hidden in the complex English description of a business challenge or the chaos of the process around you ? I am pretty sure many times. That is what I call “simplicistic intuition”. Your brain ignored all the noise, clutter and chaos to hone in on something that was the core or the gist of what was happening.

I propose that we need to think about problems in terms of “simplicistic intution” ( a term made up by me 😁). I have a lazy brain- which means that when presented with a problem, it shreds everything that is not relevant- so as not to waste even one grey cell matter (since I have so few 😊)on anything trivial . It tries to bring down everything to a state where that problem becomes something so simple that someone stupid like me can understand that. As an example, this weekend I was working on finishing my lessons on reinforcement learning for my online program. There was a plethora of RL equations presented in pure mathematical terms (but explained clearly in the videos). So first thing that I did was to draw some pictures, with zero fancy math symbols to summarize them for my dumb brain. My wife initially laughed but then I explained those boxes to her and she was impressed (finally…..after almost 13 years of marriage 😁)

But the power of translating the math to those relationships that I drew goes much beyond the coursework. If we can convert all the complexities into a view like that, it can then allow us to see what are the opportunities in the world around us to apply it. Beautiful…isn’t it ?

And this is something that I have been doing since childhood (on behest of my lazy brain). After decades of doing this (since I have pursuing academic programs kind of continuously to keep my dumb brain from becoming dumber than what it is now, since two decades), I realized few years ago that my brain has finally climbed the ladder one step above stupid πŸ˜‚. The instance when I realized this was happening was when I started deep diving into AI and ML, while concurrently being in a job where I did Operations research modeling-I had finally trained my lazy brain enough such that it had developed the “simplicistic intutition”– where if it read a business problem, a business news article,a white paper- it was able to dumb it down to my stupidity level- helping me realize what underlying solution, algorithm, methodology can be applied. Yup, it came probably a decade late than others but here it was 😊

So that was my explanation of definition of simplicistic intuition !

But while this is my stupid version, I am pretty sure that smarter people have more sophisticated versions of their own. The gist of this section therefore is- can this “habit” of simplifying things be applied to business problems ?

A2. But why is that useful or relevant ? Specifically in AI and ML context ?

The real benefit of AI and ML algorithms is that they allow you to build solutions that could not be built before. But an important aspect of these tools is also that they allow you to also build solutions around problems that we did not even try to solve before. And that is what will drive real innovation. The sad aspect is that as soon as we received these advanced tools, we started chasing the old age problems that we have solved, then solved again….and then solved some more…to solve them again using tools that have immense potential.

In today’s world, how can we develop solutions that have not existed before ? If you think about it- they have not existed before because no one has been able to see the aspect as a business problem around which we can define a tech a solution. Hailing a taxi was a pain in the a** for decades no one realized it as a problem around which we can design a tech platform solution (before Uber did).

And the reason we don’t see that is because we are trained since childhood to cherish complexity. The more complex you make things to be and then solve them, the smarter you are (has been the notion). Why do you think we have been leveraging classic operations research methods for almost half a century ? That too on the same set of problems ?

So the first aspect of leveraging simplicistic intuition is that if we develop the lens of simplifying the aspects, we will be able to find those new opportunities, hidden within all the clutter and noise. And then we can create some real value applying these advanced tools on these new opportunities.

And the second application of simplicity is in developing the solution itself. In today’s world, if you are looking to develop an innovative product, there is one aspect you HAVE TO keep in mind- someone will follow suit soon. So in order to use any innovation to your business’ advantage, you have to:

  • Build the solution fast
  • Scale the solution fast

The challenge with “build the solution fast” aspect is that most conventional businesses processes around us are not built around simplicity. Even large tech companies, that had monopolies for decades are built around ” we will roll out an update next year” model, since they had no competition. And to align everything with the pace of building a solution at ease, we put too many cogs in the wheel. (and now sudently we are realizing agile is the way to go and force fitting agile on conventional mindset and approaches). Large departments across large organizations created blankets of complexities to justify the team strength and all the pomp and show that went with it. That ” too many” also meant that to design a solution, we looked for complexity to explore to justify the headcount and salaries vs simplicity that can help us start working on something immediately.

Simplicistic intuition” allows you to find the core aspect of a problem you are trying to solve, essentially saving you significant amount of time chasing the noise and the clutter.

B. Why is simplicity underrated ?

Because we correlate simplicity with laziness. And sometimes maybe with intelligence ? It is not. Now it may stump you since I have mentioned the word “lazy” so many times in my post. Look carefully and you will see that I have mentioned “lazy brain”. A lazy brain does not want to work hard to arrive at a solution that it can get to with far less effort. But it still gets at the solution. A point to note here is that a lazy brain needs to collaborate with other brain types to create what I call “Backyard teams” (The concept is explained later in the article). The “laziness” of the brain also lets it see through noise to see opportunities that busy brains may miss, since that is what a lazy brain has been trained to do.

“Sometimes….an arsonist may be the best firefighter”.

Our systems in corporate world rewards us to solve problems that are known and have been solved hundred times before. Because we know that they can be solved and it is easier to sell those wins. If you are a new, fresh out of college supply chain analyst at a company, and you ask 2-3 problems your manager would like you to solve- they will be very focused on few areas and probe a little bit and you will find that you are not the first person solving a specific problem, and believe me, you will not be the last as well.

And remember, all this does not mean that the “simplicistic intuition” always works by itself. It means sh*t in silo. If someone did not deep dive into the mathematical nuances, there would be no algorithms around that we cherish and believe will help us identify new problems and their solutions. The intuition transforms into a solution through a collaborative effort-through teamwork.

C. Time to value Simplicity ?

Let me start this section by quoting Tom Goodwin:

“These are different times. Change requires companies to step change rather than incrementally improve. The world’s best candle makers continually made candles better but they never invented the light bulb. Today companies need to leap to new business models and rethink fundamentals and what they stand for, not slowly tweak processes incrementally.”

The unfortunate aspect is that we have designed the talent structure, processes and everything associated around those “incremental” changes. Few companies have the foresight to form teams that are “Backyard teams”. These teams will find problems that have not been solved before and by collaborating with other complementary skill sets in the same team, will develop solution around those identified opportunities, outside the crazy, execution obsessed world of supply chains. Very few organizatons have such teams. Forming and investing in such teams runs counter to the entire idea of seeing immediate benefits and results. But if you could let a team explore simplicity and then build solutions around that, just one innovative solution in a year will generate a competitive differentiator that will provide immense value to your organization.

I think I am going to stop my rant here since I am tired of typing 😁

Views expressed are my own.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s