Data Science

How Having a Blog Can Advance Your Career as a Data Scientist

This article was originally published on Towards Data Science on July 22nd, 2020.

Content creation can be scary at first. I know it was for me — sharing thoughts and opinions on subjects where there are more experienced people than me. Nevertheless, even as a beginner to intermediate data scientist (or software engineer), having a blog can kickstart your career.

I started blogging in August of 2019, roughly 4 months after starting my first job as a data scientist. That might sound strange at first because there are a ton of people out there with more experience not blogging about what they do. And I can’t reason why.

Maybe it’s a lack of time, but it probably isn’t — as publishing 2 articles per week takes around 4–8 hours for me — depending on the amount of code I need to prepare. Lack of will and motivation is a more likely scenario — mostly because people aren’t aware of the benefits of blogging.

Blogging as a side hustle has multiple benefits — and that will be the topic of today’s article. These benefits come from a single guy perspective, sure, but are easily transferable. The only problem is — blogging requires action and being able to work many hours before seeing results — something not everyone is willing to do. I’m sure you don’t fit into this category.

So without much ado, let’s get started with the benefit number one.


1 — People see what you do

Blogging is a massive reputation builder. I’m still a college student, and I had college professors come up to me and compliment my writing. And if anything — writing skills only improve with time. Getting positive feedback by someone you admire keeps the motivation levels up.

Another group of people you want to read your content are potential employers. There are two ways they can find you:

  1. By being active on the platform you write on
  2. By clicking the link provided in your CV

The first option is slightly better because they can see your progress over time. If you know your sh*t, chances are potential employers will agree with your views and opinions, making you that much easier person to hire.

From that point, job interviews should become a formality. You can’t expect this to happen if you have 3 posts to your name, sure, but you need to start somewhere. Start small, but more importantly — start today.

Having a blog also benefits you as a freelance data scientist — as it provides proof of expertise in some domain, and that’s not something an average freelancer has. As a result, you could either get more work or charge more — or both.

To summarize — everyone wants to be a data scientist. For that reason, you need to somehow stand out from the crowd. Everybody knows Pandas and Scikit-Learn, but being the guy who can teach those will work wonders for you.


2 — You learn more quickly

Data science and other tech professions are full of advanced and not-so-easy-to-understand topics. Let me show you an example.

A week or so ago I was interested in treating time series as a supervised task with Python. Doing so would allow me to use algorithms like eXtreme Gradient Boosting (XGBoost). The only problem is — absolutely no one online can demonstrate how to make predictions for new, unseen data! Guides go only as far as telling you how to forecast on the test set — and that doesn’t make the cut.

So I decided to write my own class for the task — XGBTimeSeries — based on XGBRegressor and able to handle hourly, daily, monthly, and yearly time series data, for now. It also creates different variables — like is it a weekend, did the value occur in the work hours, and many others — based only on the frequency.

And yeah, I intend to write about it — a lot. As it was not the most straightforward thing to do, and the model class is around 400 lines of code, I doubt it will fit into a single article. The more realistic scenario is that I will make a mini-course out of it. Let me know if you’d be interested in something like that.

Anyway, I digressed here a little, but I hope you get the point. Here’s a quote I like particularly:

“If you can’t explain it simply, you don’t understand it well enough.” — Albert Einstein

The quote applies to blogging, as it forces you to see things as simple as they were intended to be. There’s no point in writing about something you don’t understand because it would confuse both you and the reader.

Make sure you understand the damn thing first and then try to explain it like your reader is a 5-year-old.


3 — Money on the side

Finally, let’s talk about money. No one can tell you exactly how much you can earn, but after roughly 10 months I make anywhere between $1.5K to $3K a month — affiliate marketing included. Most importantly, blogging enables me to have options. Let me elaborate.

If you’re only working 9–5 you depend on that job. What would happen if you stopped working tomorrow? Blogging provides you options because you can have multiple streams of income if you become good enough at it — thus not being dependent on your daily job entirely.

And that’s just the tip of the iceberg. Blogging about what you do creates a personal brand around you. This won’t happen in a month, sure, but give it a couple of years. I’m not quite there yet, but I’m already seeing the impact of a personal brand in my everyday life.

“I write good articles, so readers should come to me” — is a bad attitude to have. Also, that’s not how things work. As a personal brand owner, you’re responsible for getting readers to your blog. That’s what I’ve been doing recently over at Better Data Science, and it’s not the easiest thing in the world.

Hosting your content enables you to control it just the way you want, publish your products, and promote affiliate products — all of which can result in substantial income in the long run.

Also, the “I don’t know what to write about” thing comes up a lot. It surprises me because data science evolves daily, and new things come up all the time. There are a lot of things to write about, and the list of topics will just get larger with time.


Before you go

To recap — starting a blog as soon as possible will kickstart your career — as long as your ideas aren’t completely wrong. But I trust you know what you’re talking about.

There are amazing free places to start, like Blogger or Medium — with the latter having an in-built audience, which reduces the marketing efforts. You just have to write good quality content consistently and make it appealing to the readers. There are a lot of ways to expand this, but you should always start simple — otherwise, it’s to easy to burn out.

What are your thoughts? How did having a blog affect your career? Feel free to leave your thoughts in the comment section. Thanks for reading.


Join our private email list for more helpful insights.

Dario Radečić
Data scientist, blogger, and enthusiast. Passionate about deep learning, computer vision, and data-driven decision making.

You may also like

2 Comments

  1. I really like this idea of creating blog in order to show your skills, and it realy turns me up to start creating my own blog.

    1. Do it! One year from now you’ll be glad you did it.

Comments are closed.

More in Data Science