5 Things to STOP doing as a New Writer On Medium

How to get more views EVEN IF you destroyed your algorithm…

Gabriel Klingman
New Writers Welcome

--

Me, unconvinced that I was destroying the algorithm when I first started writing on Medium

You write strong headlines.

Your content is valuable.

You’re consistent.

Yet your articles aren’t getting views… Not as many as you hoped, anyway.

You want to blame the algorithm, but what if I said you’ve trained Mediums algorithm to show your articles to the wrong people?

Mediums Algorithm: Explained

Medium wants to share your content.

Despite what the fake “growth gurus” are saying, if you create content that sparks curiosity, hooks the read, and is valuable for your target audience, Medium will share it.

The problem is that we’ve trained the Medium algorithm to share the content with people who don’t get value from it.

It doesn’t matter how valuable the latest DIY tool-review is. It’s not valuable to me so I will not click on it.

As writers, we need to train Mediums algorithm to show our writing to the right people.

The main metric Medium uses to decide whether to show your article to 5 people or 500, is the Read-Ratio.

Read-Ratio from This article by the author

The read-ratio is the amount of people who stay on your article for at-least 30 seconds (called a read), divided by the amount of people who clicked on your article (call a view).

If 100 people click (view), and 50 stay on it for at-least 30 seconds (read), you Read-Ratio is 50% (50 divided by 100 = 50%)

In my experiments, I’ve found the tipping point for your Read-Ratio is 69%. Anything below 69% doesn’t get shared, anything above 69% does.

This means, to grow on Medium you must get your article in front of the right people who click and read it.

Medium searches for these “right people” in 5 ways:

  1. Publications
  2. Tags
  3. Engagement on our article (comments or highlights)
  4. People they follow (and similar to who they follow)
  5. Based on their reading history.

This means there are 5 ways you can destroy the algorithm and train it to show your article to the wrong people (If you’re getting less than 100 views per article, this is probably why).

But this also means there are 5 ways you can train the algorithm to show your articles to the right people (and get 1,000+ views).

If you’re a non-fiction writer who’s struggling to make money, you’re in the right place.

Building a business and being great at writing are two different skills.

I send out an (almost) daily short email with business and productivity tips for writers.

Click here to join Writerpreneur.

1. Publications

How to Destroy Your Algorithm with Publications

By putting your article into a large but general publication, you are sacrificing the long-term reach of training your algorithm for the onetime reach of the publication.

Most “how to grow on medium gurus” teach that you should try to get your article into the largest publication you can find.

The publication will push your article out to a large group of people, so you’ll get a lot of views and grow your audience. Right?

Wrong.

This logic assumes 3 things:

  1. Most of the people following this publication will click on the article (they won’t).
  2. Those who click will actually read the article (they won’t).
  3. The onetime reach from the publication is larger than you can get through training the algorithm (it isn’t).

Large publications kill good articles by sharing them with an audience that isn’t interested.

You’ll get a handful of clicks, but those people aren’t hyper-interested in the content.

So they click away.

This destroys your Read-Ratio, so Medium stops sharing your article.

How to Train Your Algorithm with Publications

Publications are like collaborations: the value comes from sharing each other's audiences.

Your audience sees the publication, and the publications audience sees your article.

There are 2 steps you have to take to make this work for you:

  1. Pick a few publications that are hyper-relevant to the person you serve and/or the problem you solve.

This will get your articles into the feed and inbox of actual readers — not just random people.

These people will click and read your article.

This increases your read ratio, and Medium will share your article with more people.

2. Publish consistently in those publications.

Most people see a product 7 times before they recognize it or consider buying it.

The same principle applies to growing on Medium.

Most people will not follow you after the first read.

By publishing consistently in a few publications, you show up again and again in the feed and inbox of the same readers.

The more you show up, the more likely they are to remember you, follow you, and engage with your content.

This is how you train your algorithm through publications.

2. Tags

Imagine you walk into a room at a networking event.

In front of you, someone’s talking about the game last night.
To your right, there’s people talking about their life history.
To your left, you overhear someone humble-bragging about their job.

You decide which group you want to join based on what you want to talk about.

This is how I view tags.

A tag is a conversation you can join. To join the conversation around a topic, simply use that tag.

When you use a tag, your article will show up in the feed of those who follow the tag.

How to Destroy Your Algorithm Through Tags

2 ways to destroy your algorithm through tags:

  1. Use broad and irrelevant tags

Broad tags attract a broad audience. So using a broad tag could mean you get lots of clicks on your article.

But broad audiences have broad interests. So most of the people who click on your article won’t stay for 30-seconds (read). This kills your Read-Ratio, and Medium stops sharing your article.

The exception is if the broad tag IS very-relevant to your article.

E.g. You’re writing an article about Productivity, and you use the Productivity tag. Here, you would want to use the broad tag.

2. The Tag and the Headline not connected

Imagine sitting in a coffee shop.

You overhear a stranger say, “Wow, it’s beautiful outside.”

You interject and say, “The Boston Red Sox are gonna kill it today!”

This is jarring for everyone who hears it.

In your mind, you’re saying “The Red Sox are gonna kill it because it’s so beautiful out.” You’re adding to the conversation.

But no one else makes that connection.

You’re met with awkward stares and uncomfortable silence.

Then the stranger leaves.

This is what happens when you don’t connect your headline to the tags you use.

A tag is a conversation you’re joining. And your headline is the first thing you say when you join that conversation.

If your headline doesn't have a clear connection to the tag, then people will quietly scroll past your article.

Here’s an example of what NOT to do:
You use the tag “social media,” but use the headline “3 ways to improve your Return on Ad Spend.”

Your article could be about how social media improves your Return on Ad Spend, but the casual-scroller won’t make that connection. They will think you’re interrupting their conversation about social media, so they will ignore you.

How to Train Your Algorithm Through Tags

  1. Use relevant tags with a high follower-to-stories ratio.

Example:

“Mental Health” and “Self Improvement” tags
“Psychology” and “Productivity” tags

These 4 tags are examples of tags that have a large amount of followers (people who follow a tag) vs stories (stories that have used that tag).

The higher the ratio of followers-to-stories, the more likely people will see your article.

Just be careful that you only use very-relevant tags.

2. Make your headline relevant to the tags

If you use the tag "digital marketing" and your headline is "you're forgetting this benefit of content creation," people who follow the digital marketing tag won't click on that article — even if the article caters to them.

When they see the article through the tag “digital marketing”, their focus is on digital marketing.

Not content creation.

You could have the same article with the headline, “cut your digital ads cost in half with this trick”.

This article's packaging applies to the people following the "digital marketing" tag.

When someone sees it in the “digital marketing” tag, they’re more likely to click on it because it connects with the frame of mind that they are already in.

3. Engagement

Destroy Your Algorithm Through Engagement

Engaging with other people’s article via “clap for clap, comment for comment, read for read,” etc. will destroy THEIR algorithm.

Medium now recommends content to you based on your behavior and the reading behavior of others like you.

So if you do clap-for-clap, comment-for-comment, or read-for-read with people who are not your target audience, Medium will start showing your article to other people like that person (not your target audience).

These people (who are not your target audience) will see the article and won’t click on it — or worse, will click on it and won’t “read” it.

This tells to Medium that your article is poor quality and they’ll stop sharing it.

Train Your Algorithm Through Engagement

There are 4 ways to train your algorithm through engagement:

  1. Use a call to action to engage (e.g. comment, highlight, save, etc.).

The more people that engage, the broader reach your article will have — because it will get shown to the followers of those who engaged.

This also means the quality of the person who comments and engages is crucial — which leads into the second strategy:

2. Consistently leave authentic and valuable comments on the articles written by the top-writers who are currently serving your target audience.

Most people don’t realize this, but if you show up consistently for a creator (responding to their emails, commenting on their articles, and leave genuine feedback or value), they will remember your name.

Do this enough and they will probably follow you

If you create articles that are interesting to that creator, then they will probably click on and either highlight or comment on some of your articles.

This will push that article out to their audience.

3. The third strategy is by taking a strong public stand for something.

If you have strong values and beliefs (especially if they’re controversial), take a public stand for them.

Here are some of the public stands I have taken.

  • I will not join the Medium Partnership Program — and I will make more money this way.
  • I f*cking hate subscriptions.
  • You don’t need another f*cking course. You need community.
  • If you want to build a business through writing, stop working on your “big project” (book, course, coaching program, etc.). Before you focus on monetization, you need to capture attention.
  • I will not follow you or comment on your content if I’m not your target audience — because I don’t want to f*ck up your algorithm.
  • The value of an article comes from the idea, not the grammar or spelling.

These are controversial for my audience.

Each time I say them on Medium, the comments flow in.

There is an important key here: you must actually believe what you say.

If you say something controversial for the sake of being controversial, that’s manipulative. The audience will catch on and stop following you.

4. Use controversial language

If you say something that gets an emotional reaction, you’ll likely get a comment.

I’ve used this tactic in this article where the opening line is, “Unless you’re Elon Musk or Trump, no one cares about your opinion.”

And later in the same article, “if you’re not genuinely giving value in every sentence you write, you are robbing people of their attention.”

These are strong stands, presented controversially.

And yet… Over 200 people joined my email list from this article alone.

4. Follows

Destroy Your Algorithm Through Follows

There are 2 ways gaining followers can destroy your algorithm:

  1. Writing content that’s too-broad.

When you write broad content, you reach a wider audience. If the content was valuable, you gain followers.

This sounds great, but there’s a big problem lying under the surface….

Medium recommends articles to you based on who people similar to you also follow.

So if you write broad content that doesn’t connect to the problem you solve, you gain followers who may not be your target audience.

So if you gain a lot of followers who are not your target audience, Medium will start recommending your content to more people who are not your target audience.

When those people don’t follow you, Medium will assume the quality of your content is poor and will recommend you to fewer people.

Here’s a video showing how this same principle works on YouTube:

2. Follow for Follow

This causes people to follow you who aren’t your target audience.

Medium will start suggesting you to people who aren’t your target audience.

They won’t follow or engage, so Medium will assume the quality of your articles is poor and will stop suggesting you to people.

Train Your Algorithm Through Follows

“Followers are just a vanity metric, right?”

WRONG.

When someone follows you, it’s the same as if they are buying stock in the content you create. (Thanks Martin Messier for the metaphor here).

What a follow communicates is, “What you wrote today was valuable and I believe what you will write tomorrow will be more valuable. So I will follow you today so that I can see what you write tomorrow.”

A follow is a validation that what you write has value for a specific group of people.

How can you use a follow to your advantage and train the algorithm?

By using your follows to improve your skills.

You can follow people who are hyper-relevant and have your audience.

By following them, you’ll see what type of content resonates with your audience.

The headlines they are using, the content structure, the ideas they are communicating.

This will help you in your own content creation process.

5. Reading History

Destroy Your Algorithm Through Reading History

Medium recommends articles to individuals based on their reading history.

If I read an article about Buddhism, Medium will recommend more articles about Buddhism.

This is a great way to grow organically, but it causes a problem when people who are not your target audience read your articles.

This happens when you write articles that are…

  • Overly-broad
  • Unrelated to your target audience
  • Exchanged in read-for-read practices

In these cases, someone who’s not your target audience will read your article.

Medium will see this as a good sign and push your other articles to this person.

When they ignore them, Medium will assume the quality is poor and stop recommending your articles.

Train Your Algorithm Through Reading History

You can train the algorithm to show your article to the right people by getting your target audience click on and read it.

The easiest way to get your target audience to click on and read your article is to craft headlines that attract your target audience and repel everyone else.

  • E.g. Look at the headline of my top performing article.

It calls out the target audience (new writers) directly in the title.

This is one of the simplest (yet strongest) tactics for getting your target audience to click on your article.

If you’re a non-fiction writer who’s struggling to make money, you’re in the right place.

Building a business and being great at writing are two different skills.

I send out an (almost) daily short email with business and productivity tips for writers.

Click here to join Writerpreneur.

--

--

Gabriel Klingman
New Writers Welcome

Ops Manager for Capitalism.com. In March, I wrote 70k words in 7 days. Follow to learn the business of writing.