Premature Optimization

“What do you want to be when you grow up?”

You’ve probably heard that a lot as a kid.

With a few exceptions, if you really think you had the answer to that question, you’re probably making the mistake of premature optimization.

Googling “Premature Optimization” will tell you – “Premature optimization is the act of trying to make things more efficient at a stage when it is too early to do so”.

What it is

Premature Optimization is the attempt to try and get it right at once, to figure all the little details beforehand, before you actually get started.

Trying to micromanage, to “foresee all eventualities”, to have a fixed, well thought-out schedule beforehand.

It’s the tendency to have a static plan, incapable of being modified. To stick to the script like a robot.

Or it could be the fear of getting started at all, because you keep thinking you don’t know how to reach the end.

Where you see it

It’s everywhere.

It’s actually supposed to be very impressive.

It’s “someone who knows what they want”. The person who’s completely sorted and certain about their choices, without an iota of doubt.

These sort of stories make for very good reading in hindsight – a young lad determined to be an entrepreneur or a small child clear that they want only to be a civil servant and nothing else.

I usually take them with a bucket of salt – it’s typically the product of someone spinning a nice story hoping it’ll sell. But even if true, it could actually be more a case of parochialism and limited outlook than of clarity. It might just happen that that person ignored all other possibilities in life, unless of course they actually evaluated their options and settled on the best one, which is commendable. But regardless, there’s no reason to try aping them.

In Personal Life

Thinking you’re completely sure about what you want, about your preferences – even without having complete information.

Believing that you want to buy a laptop of X brand or shoes of Y brand – without even checking out what’s available. There’s a chance you could find something better – at the very least, you might want to be open to that possibility instead of shutting it off at the outset.

Or spending lots of money on equipment for a new hobby – then realizing you don’t really like it that much.

Deciding what you want to be when you grow up when you’re still in school. Again, you might go ahead with that plan, or you might change your mind. But fixating on it and shutting off other possibilities is stupid.

Deciding what sort of people you generally befriend, or whether a stranger you just met is worth befriending. Using pre-conceived biases to evaluate people and perhaps write them off without a chance. Again, you might think you only get along with introverts but it could easily happen that the extrovert you run into becomes your best friend if you get to know him.

In Academics

A lot of people ask for time-tables, schedules, and above all else, “the unique strategy” for clearing UPSC or other competitive exams.

One reason it’s pointless to ask this is that what works for you won’t be the same as what worked for someone else; you’ll have to put in the effort to figure out what floats your boat – discussed here.

But apart from that, you can’t possibly figure out how to do study for such an exam or how much time to spend on each subject – until you get involved in the process.

You just don’t know enough about a field until you start doing it. Someone who only reads about war probably doesn’t know how to fight, someone who only reads about singing probably doesn’t know how to sing well.

In Business

Here, it’s the belief you know exactly what customers want and thus spend loads of time and money making a perfect product – sometimes only to find no one wants it.

That’s why Minimum Viable Products (MVPs), iteration and “failing fast” are often helpful – make something, learn more about what people want, and tailor the plan accordingly.

And as you go along, you get new ideas and modify old ones. It’s doubtful the biggest firms of today like GAFA ever started off knowing exactly what they were going to be today. And it’s also likely they don’t know what they’ll be like in a decade from now.

Sticking to a pre-decided plan almost guarantees failure. Reality never matches up to our script, and that means you can’t handle the challenges you’ll face.

How you fight it

Overcoming Premature Optimization doesn’t mean going without planning or being completely spontaneous, taking things on the flow.

You can, and probably should, still plan in depth.

Going into something without thinking about it is setting yourself up for failure.

It simply means being open to the possibility – and most probably necessity – of changing your plans and modifying them to adapt to circumstances or new learnings.

How do you actually change your plans though?

It can’t be blind. It should be based on data or new information, but the question always arises – which information? You need to decide what’s relevant.

Maybe someone sets out with a time-table for the civil services exam and devotes 3 hours a day to History and 3 to current affairs and only 1 to Polity. After giving a few mock tests, if that person scores low in Polity and high in the rest, it’s a clear signal to re-allocate his hours between his subjects.

But changing your optional simply based on the “trends” in marks in one year isn’t necessarily a good idea – there’s no reason to think the so-called trend will hold the next year.

Or, a company allocating its marketing budget equally between Twitter and Facebook and Instagram might find that Instagram leads to the highest revenue per dollar spent, and spend more there instead.

What does it take?

Courage – to take a leap with imperfect knowledge. To be able to go into something like a business or an exam without knowing everything about it in advance but not letting that paralyze you, because you’re confident you’ll figure it out as you go along.

Humility – to accept that you won’t have all the answers right at the beginning. Without humility there’s no way you can learn what you need to know because you won’t even try. And without humility you can’t abandon what you were doing when it’s clear it doesn’t work, because that means accepting that you were wrong.

Wisdom – To know that you don’t know (a lot) yet. It takes a little intelligence to recognize what doesn’t work, and probably a lot more intelligence to understand why it doesn’t work. Only when that’s done can you move towards figuring out what can work.

Summing Up

Understanding Premature Optimization means you don’t try to solve everything right up to the end.

If it’s like running a 1000 metre race, then solving the first 100m and having a slightly vague idea about the next 100, an even more vague one about the next 100, and so on, should probably be good enough.

Maybe this doesn’t apply to thing like space launches or setting up nuclear reactors – but these are different situations, I think.

Why are they different? Firstly, the cost of failure is so high that you don’t mind redundancy, spending lots of extra time and money to reduce the chance of failure even slightly. The higher the cost of failure, the more you’d plan in advance to mitigate it.

Secondly – they’re slightly more predictable environments. It’s possible, though not easy, to model the trajectory of a spacecraft or the threats to a nuclear plant to a large extent of accuracy. But you can’t predict customer preferences or new entrants in a market with the same accuracy, because unexpected events that can’t be modeled happen more frequently.

So what it means is – do your planning thoroughly but don’t expect to be finished with it.

Don’t delay execution because the “strategy” isn’t 100 percent ready yet.

It could be 20 or 30 or 50 or whatever you’re comfortable with and what fits your time constraints. The rest is impossible to plan as of now – the path only becomes visible as you move ahead.