5 Habits That Made Me A Data Scientist

Author:Murphy | View: 25255 | Time: 2025-03-22 21:02:24

Becoming a data scientist is not easy, but some essential habits helped me in my journey, which is what I will share with you today!

Hopefully, these habits are something you can pick up along your process, as I think they will make a great data scientist in the long term.

Action-Oriented

At the start of learning data science, just get stuck straight in. Stop thinking and start doing.

Not overthinking things and starting learning immediately is the best way to go and everyone pretty much says this.

For example, I always mention that there is no such thing as a "best course." Sure, some tutorials may be better than others, but any entry-level course generally covers the same content.

Instead of spending hours pondering which course to take, imagine the progress you could make if you invested that time in actual learning and studying the material.

Similarly, don't waste time trying to find the "perfect project". Just find one you like and get going; you can always start a new one later or change your current one. Learning is what you are after, not a project outcome, especially at the start of your journey.

The primary advice I recommend for being action-oriented is to time block "thinking" and "research" time to avoid spending too long on it. This way, at the end of "scoping", you know you need to start working on something, even if you haven't done all the background work.

Consistent Learning

The most important habit is consistent learning. You only get better through continuous practice and doing a bit each day. I am sure you have all by now heard that improving 1% every day leads to exponential growth.

1% improvement every day is 1.0¹³⁶⁵ ~ 38
1% decline every day is 0.9⁹⁹³⁶⁵ ~ 0.026

It is a cliche but very accurate, and you should leverage this to your advantage as much as possible.

Every day, I dedicate time to learning new topics or improving my skills. That could be studying ML theory, challenging myself to a Coding problem, or even reading an article online. It can be really simple, so don't overcomplicate it.

Every little bit you do makes a difference, so try to carve out time each day, even if it's just half an hour.

Some advice I recommend for this habit is:

Time block learning and studying periods in your day. A common thing I do is use my commute time as learning time.
Get subscriptions to platforms like Medium and newsletters and follow Data Science and machine learning YouTube channels.
Keep a consistent backlog of things you want to learn so you always have ideas.

Goals & Roadmaps

The previous habits are very important; however, without a clear roadmap or goals, they may lead you slightly astray as you have no clear direction.

It would help if you tried to have a roadmap or at least an end goal of what you are trying to achieve. For example, sometimes my goals are as simple as "learn neural networks and implement one in PyTorch." Realistically, this is quite vague, but it gives me something to aim for while studying and learning.

Roadmaps are better as they are like study syllabus, similar to what you get at university. They will structure your learning and ensure each part follows on from the last one. They typically also contain the resources you should use, so this saves you time in this space as well.

I followed a roadmap while learning data science and technical areas like time series and deep learning. This greatly expedited my progress, as I didn't waste any time thinking, "What's next?" or "How do I learn this?" Everything was self-contained, so I could purely focus on the material with limited distraction.

There are loads of roadmaps out there, and I have even created some myself, which you can check out below:

How I'd Become a Data Scientist (If I Had to Start Over)

How I'd Learn Machine Learning (If I Could Start Over)

My main advice for this habit is:

Have a clearly defined end goal you want to achieve.
Get a roadmap; pretty much any would suffice.

Stop Comparing Yourself

A common problem in today's society is comparison. Due to the online world, we are exposed to so many people, and often, we only see their highlights because people rarely post their failures (I am guilty of this too).

Not to mention, you will also see everyone who is ahead of you, and as humans, we tend to compare upwards and never look back to see how far we have come.

Without getting too cringy, you should only compare yourself against yourself. Everyone else is redundant, and you should always focus your attention inward.

Take a leaf out of Matthew McConaughey's book. When he won Best Actor at the 2014 Oscars he said that…

My hero is me in 10 years.

This is the kind of self-belief and self-improvement mindset we should all strive for. It's you vs you always.

Instead of worrying about others, focus on your own development. It may take you longer or even shorter, but that doesn't matter. What matters is that you are making progress. This progress is made much quicker when you stop worrying about everyone else and it makes it more enjoyable as well.

Data science is a field full of some of the most innovative, intelligent, and hard-working people. Chances are, you won't be the best data scientist in the world, and there will always be someone better than you.

This is not to discourage you but to show you that comparing yourself is futile, as you will never win. So, like I just said, always look inward and look to improve relative to yourself.

My main advice is:

Avoid doomscrolling about other people and their accomplishments.
Try and be inspired by other people instead of being envious.

I appreciate that these are pretty hard to do as they require a mindset shift and are not as straightforward as the ones above.

Patience

Remember, great things take time. In the case of data science, it's crucial to be patient and embrace the learning process as a journey, not a race.

Reflecting on my own journey, I've come to accept that mastering a complex field like data science is a long-term commitment. Despite working as a data scientist for almost three years, I still feel like I'm only scratching the surface.

Many videos and articles online say you can learn data science and get a job in 3 months. This may be true for some people with significant advantages like previous coding knowledge or a PhD in a STEM subject, but this won't be the case for many others.

So, if you are a couple of months into your roadmap and you are struggling and finding the information just won't sink in, give it time. There are literally three-year full-time bachelor degrees in data science now, which shows the type of time scale you should think in when learning a field like data science or Machine Learning.

For example, I am currently looking to learn more about the engineering and deployment side of data science and machine learning. I am under no illusion that this is probably going to take me a good few years to master; I am comfortable with this.

My main advice for this bit is:

Give yourself time and set more extended deadlines than you may need.
Relax and enjoy the journey!

Summary & Further Thoughts

To quickly recap the habits

Action Oriented: Do more and think less
Consistent Learning: Try and learn something every day
Roadmaps & Goals: Have clear goals or follow a roadmap/syllabus
Comparison: Compare yourself to yourself and no one else
Patience: Learning data science takes time; think in years rather than months.

I hope you can adopt these habits. They have transformed my abilities as a data scientist, and I am sure they will do the same for you.

Another Thing!

I have a free newsletter, Dishing the Data, where I share weekly tips and advice as a practising data scientist. Plus, when you subscribe, you will get my FREE data science resume!