How I Made a Massive 4000 Word Article About Math the No.1 Trending on Medium

And how can you do it, too

There are three things I am incredibly passionate about in my life: mathematics, machine learning, and teaching those to others. One form of expression for this is writing. I love writing: I feel like I am capable of taking complex ideas and explain them simply. I have learned this skill during five years of teaching mathematics to university students.

In the past year, I have been publishing on Medium consistently, with varying success. On average, my posts attracted around 10,000 views. Some reached 50,000 after a few months, some are still stuck below 1000.

Last week, I published my 38th article on Medium. It was a massive 4000 words long reading, laying a roadmap towards understanding the mathematics behind neural networks from the ground up.

What happened was beyond my most optimistic expectations.

37,000 views in the first day, almost 100,000 views in the first week. 2,500 claps and more social media shares than I could count. The post became No. 1 trending on Medium and stayed there for a few days, overtaking Greta Thunberg’s story. (Although she had much more important things to say.)

However, this insane success was not an accident; it was engineered for the most part.

Everyone wants to go viral, but few do. Most writers tend to churn out articles and apply “growth hacks” to make it. I won’t pretend I haven’t walked this path before. Paradoxically, the key to viral success is not a quick hack but rather an insane amount of hard work.

When everyone compromises quality over quantity, you must do precisely the opposite: double down and create the best content there can be.

How does one make an article about mathematics trending?

When I started to write that ominous and insanely successful article more than a month ago (yes, it took me that much time), I wrote down three rules for myself. During the creative process, these were my guiding principles.

What were they?

1. Write the single best resource on the topic.

If you are a technical writer, you know that the competition is enormous. Throw a stone, and you’ll hit a data science content creator. Just on Towards Data Science, there are a ton of new posts every day. There are a million “Get Started With Pandas” pieces out there.

Every single niche topic has dozens, if not hundreds of articles about it. If you don’t stand out, you will disappear in the mass.

You want your article to be the best among all others. How do you achieve that?

• Pay attention to all the details, and don’t go only for the low hanging fruits. Go deep into concepts that are not unraveled in other posts.
• Take the time to really understand things. Read research papers if you have to, then explain the most difficult technical details in a way such that even newcomers to the field would get the gist.
• Don’t be stingy with time. I have spent at least a month writing that explosively successful article.
• Never compromise on quality. If you think that something can be done better, do it.
• Sharpen your skills and make your content exciting to read. Pay attention to grammar as well. Broken English makes the readers instantly close the article, no matter how attractive the topic is. I use Grammarly, which helps me with both. Even the free version can go miles for you and massively enhance your posts.

Besides the quality of your writing, another essential contribution to the story is the topic selection. This ties into the next rule I set for myself.

2. Don’t create content with an expiration date.

One apparent trend you can instantly observe is that only a tiny percentage of all content will stay around for a long time. Articles like “Top 3 Awesome Features of PyTorch 1.7” or “10 Books About Data Science To Read in 2020” have an expiration date.

No one will be interested in a post about PyTorch 1.7 in a few months when 1.8 is out. No one will read a listicle about “books to read in 2020” in the following years.

Sure, these can get decent views in a short time (but only if you are the first to write about the topic), but will it establish you as an expert in your field? Will people find you based on them in a few years and connect with you professionally?

No.

Think about the book Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville. Written in 2016, it is of the most referenced textbooks in the subject, and it will be one for many years.

Strive to be like this in an article format.

To set a good example, check the Object Detection for Dummies series from Lilian Weng or Deep Learning, NLP, and Representations from Chris Olah. They are three and six years old, respectively, but I still revisit them occasionally — impressive writing quality, combined with an exciting and timeless topic.

Work hard to make every article of yours like this.

3. Create a beautifully designed visualization and jam-pack it with knowledge.

What would you rather click on? A block of text from an interesting topic or a stunning graphics?

Research and empirical evidence shows the latter. Most people are visual, and it is easier to facilitate engagement with images than with words.

One way to quickly convey in-depth knowledge is to package it into beautiful infographics. I have learned this from Neil Patel, one of the top online marketers of our time. Although blogging about machine learning is admittedly a different art than online marketing, some of the same principles hold: good infographics captivates people and increases engagement.

The one in my post was shared on LinkedIn and Twitter more times than I would dare to count. The visuals made people click, the content made them stay and share.

If you don’t have any visual skills, you can hire a graphical designer on Upwork or Fiverr to turn your sketch into a captivating illustration. (This is what I did, for example.)

Conclusion

If you are a technical blogger, pumping out mediocre articles is not the way to go. In my experience, the road to success is contrary to popular advice.

Be the writer you would want to read.

Create the best content among all the others by focusing on quality.

Write about impactful topics, not rapidly fading trends.

Make stunning visualizations.

Stop growth hacking. Focus on the substance.

Thank you for reading! My name is Tivadar, and I am passionate about democratizing machine learning. If you would like to catch my latest thoughts, you can catch me on Twitter @TivadarDanka, where I ponder about deep learning, developer life, and building a startup to revolutionize crowdsourcing for machine learning. (And Fight Club occasionally.)