Why you shouldn’t be a Perfectionist as a Data Scientist

Admond Lee
Jul 23, 2020


I’m not sure about you.

Maybe you’re a realist who always thinks realistically and acts in a practical way given any situation to you.

Or maybe you’ve struck a balance between being realistic and chasing perfection at the same time.

But chasing perfection has always been something that I’m guilty of.

I’m not saying that chasing perfection is a bad thing, but chasing perfection “could” be a bad thing, especially when you’re in data science field.

“Perfection is not attainable, but if we chase perfection we can catch excellence”
— Vince Lombardi

After being quite a while in data science field upon my transition from academic to data science path, I’ve learned so much when it comes to being able to deliver business value while still having the bandwidth to make an excellent deliverable.

Hence I hope to share my thoughts and experience with you to hopefully give you the awareness in your day-to-day data science work and how to overcome the perfectionist residing in you.

By the end of this article, I hope you’ll understand better why chasing perfectionism might not be a good idea as a data scientist and how to overcome that in your career — or perhaps, in life.

Let’s get started!

Why you shouldn’t be a Perfectionist as a Data Scientist




You see.

When you’re a beginner in data science field, it’s easy to fall into the trap of chasing perfection blindly without understanding the real needs of the business in your company.

Imagine you’ve just joined a company as a junior data scientist, and now you’re given a project to build a deep learning model for image classification and put it into production in six months.

Needless to say, you’re so excited about the project and already bouncing some ideas around with your team members.

Six months after, you miss the deadline for final deliverable as you still think that your model is not good enough, despite the fact that your model has already outperformed the existing baseline model.

So you keep tuning your model to hopefully squeeze the last bit of improved accuracy out by 2%…

And because of this, you miss the deadline for the project deliverable (putting the model into production) and your boss is skeptical of your credential and capability as a data scientist. What happens next? Now you become frustrated and wonder why your boss simply doesn’t understand your work.

Does that sound familiar to you, somehow?

“Done is better than perfect”

In fact, the similar case happened to me when I first got started in a data science internship last time.

I read a lot of research papers.

I tried to train and tune different kind of models to find the best one without having a full understanding of the business needs.

What I knew was that I got to get the best model ever like what I did in any of the Kaggle competitions.This is exactly the difference between Kaggle competitions and real life business projects.

In Kaggle, you’re your only stakeholder and you just have to train the best model and “boom!”… You win the competition.

In real life business projects, this is not the case. You have your bosses and other stakeholders with limited resources (time or $$) to deliver the final results.

And now, what I’ve realized is that delivering real business value is the ultimate objective in any data science projects. And if you’re a perfectionist, chances are you can’t always deliver solutions to your customers because you might fail to deliver Minimum Viable Product (MVP)to them.

With MVP and feedback from your customers, you can then further iterate and improve the future product development (or data science solutions, whatever name that you may call it).

If you’re a perfectionist
, you’ll always tend to set unrealistic expectation of yourself and everyone.

If you’re a perfectionist
, you’ll find yourself struggling to get things done (or you may think your work is always not good enough to be considered done)

If you’re a perfectionist
, you’ll be afraid of making mistakes and that deprives you of any opportunity to learn and improve.

How to overcome the perfectionist in you


Knowing when NOT to be a perfectionist
could sometimes be a blessing in disguise.

To be honest with you, I admit that the perfectionist in me still exists. And I think this is still a good thing, if and only if I know how to go for perfection “timely”.

That being said, here are some of the steps that I personally use whenever I realize that I tend to procrastinate (I mean to go for perfection).

1. Be aware of it

Yes. Self-awareness is key here.

If you’re not even aware of the moment when you start going for perfection, there is no way that you can overcome that. In other words, to overcome the perfectionist in you, first you need to be aware of that.

Start questioning yourself why you’re doing what you’re doing and what you’re doing will truly benefit the business value. By having this inner conversation with yourself, you’ll begin to prioritize the right things at the right time.

I know this sounds a bit abstract. So take your time to digest and practise.

2. Challenge yourself to establish more realistic expectations of yourself

Create a Minimum Viable Product (MVP) and understand “what’s good enough” and deliver that to your customers.

Depending on what problems and industry that you’re solving, there are certain areas where less-than-perfect is perfectly acceptable.

Be prepared to set that realistic expectations of yourself instead of going down the infinite rabbit hole in the name of deeper investigation and making results better. Nevertheless, the ideal case is always to go deeper and make results better (or more accurate), provided that time and other resources allow.

3. Accept mistakes and failure

There is no absolute perfection. Because there is always room for improvement.

Embrace failure and the fact that you’re going to make mistakes along the way. Just be resilient and learn from your mistakes to improve.

This way you’ll be able to overcome the perfectionist in you by constantly making progress towards your goals and improve along the way.



Final Thoughts



Thank you for reading.

I hope by now you already understand better why chasing perfectionism might not be a good idea as a data scientist and how to overcome that in your career.

While chasing perfection is perfectly fine, it’s up to you to decide and prioritize what’s the most important task to do, depending on your situation and context.

As always, if you have any questions or comments feel free to leave your feedback below or you can always reach me on LinkedIn. Till then, see you in the next post! 😄

About the Author
Admond Lee is now in the mission of making data science accessible to everyone. He is helping companies and digital marketing agencies achieve their marketing ROI with actionable insights through innovative data-driven approach.

With his expertise in advanced social analytics and machine learning, Admond aims to bridge the gaps between digital marketing and data science.

Check out his website if you want to understand more about Admond’s story, data science services, and how he can help you in marketing space.

You can connect with him on LinkedIn, Medium, Twitter, and Facebook.

“Why you shouldn’t be a Perfectionist as a Data Scientist”
– Admond Lee twitter social icon Tweet


Share this article:

1 Comments
  1. ds1
    ds1
    6 months ago
    Great article!

Post a comment
Log In to Comment

Related Stories

Jul 23, 2021

Pandas vs SQL. When Data Scientists Should Use One Over the Other

A deep dive into the benefits of each toolTable of ContentsIntroductionPandasSQLSummaryReferencesIntroductionBoth of these tools are important to n...

Matt Przybyla
By Matt Przybyla
Jul 14, 2021

How To Write The Perfect Data Science CV

These tips are also applicable to Software Engineers. Make a few changes in your CV and land that job!Writing a good CV can be one of the toughest ...

Roman Orac
By Roman Orac
Jul 09, 2021

Separating Hype From Value In Artificial Intelligence

You've probably heard a lot about data science, artificial intelligence and big data. Frankly, there has been a lot of hype around these areas. Wha...

Daniel Morales
By Daniel Morales

Win USD $2,000 in cash prizes with our data science competition!

🎉 Model submissions for the "Keyword Recency Prediction" competition will close in

arrow-up icon