avatarRené Junge

Summary

The author's experiment confirms that publishing new content significantly increases overall earnings on Medium, despite the new articles themselves attracting little interest.

Abstract

The author conducted an experiment to determine if publishing new content on Medium affects earnings, even if the new content does not generate substantial interest. The results, gathered from the first of December, indicate that daily earnings increased by over 64% on the day a new article was published, compared to the average daily income in November. This increase is attributed to a surge in readership of older articles rather than the new content itself. The author's thesis is supported by the observation that earnings drop when no new content is published, reinforcing the idea that Medium's algorithm favors accounts with recent activity by promoting their entire catalog of articles.

Opinions

  • The author believes that Medium's algorithm rewards the publication of new content by increasing the visibility of an author's older articles.
  • The author expected at least a 20% increase in daily earnings with the publication of a new article but observed a much larger increase, suggesting a significant impact of new content on overall earnings.
  • The author notes that the increased earnings are not primarily due to the new article's performance but rather the renewed interest in older articles.
  • The author's thesis includes two premises: earnings are higher on days with new content, and this increase is not solely due to the new content's performance.
  • The author is curious and open to further validation of their thesis with future publications and their corresponding effects on earnings.

Yes, the Algorithm Really Rewards New Content

Here is the result of my experiment from the first of December.

Photo by Brayan Espitia on Unsplash

In my last article, I asked the following question. Do your earnings increase as soon as you publish a new article, even if this article only attracts little interest?

Today, six days later, I can answer this question with all due caution with yes.

As a reminder, here again, is the article in which I present my thesis and the experiment designed to test it:

As we recall: in November, my average daily income from Medium articles was $2.65. I published only eleven articles in that month, and the last article in November came out on the twenty-fifth of November.

So, when I published the above article, nothing new had been posted by me for five days.

Since I expected a new article to increase my earnings by at least twenty percent on the day of publication, I should therefore make at least $3.18 on the day of the new article.

Then, on the second of December, I had the figures for the previous day. I had achieved daily earnings of $4.41. That was over 64 percent more than my daily average in November.

Twenty percent, which I had expected, could still be interpreted as a normal fluctuation, but we can speak of statistical significance at sixty-four percent.

The new article itself, however, had little part in this result. Only eighteen cents of the $4.41 went to the account of this article.

Thus I see the two premises of my thesis confirmed.

Premise 1: On days with a new article, earnings are significantly higher than those without publication.

Premise 2: This higher earnings results for the most part not from the new article itself, but from older articles that suddenly find more new readers again.

Another indication for the correctness of my thesis: Since the first of December, I have not published another article, and as a result, my daily average has dropped again to 2.19 dollars.

I am curious to see if my article today will again prove my thesis.

Read every story from René Junge (and thousands of other writers on Medium).

Your membership fee directly supports René Junge and other writers you read. You’ll also get full access to every story on Medium.

Just click here and get tons of exiting and fresh content every day.

Writing On Medium
Writing
Blogging
Experiment
Publishing
Recommended from ReadMedium