Artificial Intelligence on how to succeed on Medium

Image by fauxels from Pexels

Every time when I look for any quality content on the web it usually brings me to articles on Medium. I’m mainly interested with publications on data science, artificial intelligence or programming where the quality matters. From my point of view, quality authors with quality articles have turned Medium one of the most successful content publishing platforms on the Internet. It offers lots of knowledge and, in the era of remote work, many possibilities… well, no wonder it has attracted so many writers.

But, how to take advantage of these opportunities as a writer? How to succeed and become a top author on Medium?

Curious to know what it takes to make articles popular I programmed and trained an “AI” algorithm called LexRank to find out the answer. In short, LexRank is a Natural Language Processing (NLP) model that is able to analyze the huge amount of text and extract the most vital phrases (theory with examples can be found here).

Let’s see results!

Key findings given by the model

I discovered that sentences highlighted by the model can be grouped into three main topics concerning stories, social media and being patience.

Writing isn’t easy, and writing a good story is even harder. That’s why I post mostly technical articles :). Nevertheless, it seems that being yourself and writing about an engaging and truly story/topic is one of the most important things to get a wider audience. Below are a few highlighted phrases by the model concerning stories:

  • “A great blog post isn’t the same as a great story”
  • “It’s a personal story”
  • “Be Yourself”

Don’t forget, at the end of the day, all this works just like with many other social media platforms. This is what I got from a few of the most important phrases highlighted by the model concerning social media:

  • “Everything you’re doing to grow your Twitter and Facebook following will ultimately help you on Medium”
  • “Promote Medium posts to your email list”
  • “Get users attention so that they follow you on social media and start looking out for your content”
  • “Make friends on Medium”

Rome was not built in a day. More haste less speed. Slow and steady wins the race — old proverbs, but still in use. This is based on the next few highlighted phrases by the model:

  • “Realize that it May Take Time”
  • “Focus on Quality”
  • “Keep Learning”

Now the question is — can the model be wrong? Off course, it can be wrong! It was trained on the data created by other writers. So don’t take it for granted and just keep writing, learn and have fun!

Thank you!

Extras: python code for training your own model

The below chunk of code can be reused to generate similar conclusions. The only thing you have to do is to collect articles on “how to be a top writer” or “how to succeed on Medium” and put them into DATA_FOLDER as a separate files.

Note: you can find almost the same chunk of code here, where I used AI to check on how to lose weight.

from lexrank import STOPWORDS, LexRank
import os
from path import Path

DATA_FOLDER = 'succeed_on_medium_data'

documents = []
data_path = Path(os.path.join(os.getcwd(), DATA_FOLDER))

for file_path in data_path.files('*.txt'):
with'rt', encoding='utf-8', errors="ignore") as fp:

all = []
for doc in documents:
for sentence in doc:

lxr = LexRank(documents, stopwords=STOPWORDS['en'])

summary = lxr.get_summary(all, summary_size=20, threshold=.1)

for phrase in summary:

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store