A Medium Experiment

Writing Through the Rain

Last week I decided to try out cross-posting a piece, “The Hollywood Personal Egg Service That Wasn’t,” to both the candler blog and to Medium. I had only ever posted one other piece on Medium, and I thought it was time to give it another shot. Here’s what I learned.

The Setup

I write everything for the candler blog in Markdown using Ulysses III on the Mac. (I’m also beta testing the forthcoming Ulysses for iPad, which is incredible.) Then I do my final edit and formatting in TextMate 2 and check the piece in Marked 2.

In order to get my article into Medium, I copied the formatted piece from Marked 2 and pasted it into a new Medium draft. Then I do some cleanup (Medium always adds a trailing line after block quotes, as one example) and add any images.

The only difference between the two pieces I published is the number and placement of images. I added more photos to the Medium piece as well as captions. I did this to take advantage of Medium’s design elements. If there’s anything that attracts me about Medium it is the way it displays images and other media.

I published the candler blog piece a few hours before the Medium piece. I posted about it on Twitter and Facebook and emailed it to Jason Kottke. For this experiment, I didn’t promote the Medium piece at all. I wanted to see what would happen when a piece is thrown into the wilds of the site.


Jason was kind enough to post the candler blog piece to his recently revived “Quick Links” section (formerly “remaindered links”), which pulls in articles from his Twitter feed and publishes them to the front page of kottke.org. Almost all of the people who found the piece found it through Kottke’s quick links.

Meanwhile the Medium version of the piece went unread for a day until the “Medium Staff” account “recommended” the piece. As far as I can tell, this meant that anyone following “Medium Staff” would see it in either their main “Home” feed or under the “Culture” tab. “Medium Staff” is basically a default account for new users, kind of like Myspace Tom. It has over 1 million followers (or exactly 1 million; Medium obfuscates big numbers). Five other accounts recommended the article. I also garnered one new Medium follower along the way.

The Numbers

After almost a week online, the candler blog article has been viewed 1,863 times. In addition, I have a little over 1,000 RSS subscribers, about 200 of whom opened the piece in an RSS reader.

The Medium version of the piece, meanwhile, has 1,240 views and 300 “reads.” Medium puts a lot of value on “reads” and “read ratio.” It’s unclear exactly how they calculate how many people actually read a piece, but I’m guessing it’s either based on user time spent on the page or scroll position, or both. That gives me a read ratio of 24%.

How many people who came to the candler blog actually read the article? While an imperfect measurement, my Google Analytics tells me the average time on the page with the piece is 3 minutes 31 seconds. That’s enough proof to me that at least one quarter of the visitors, and probably more, actually read the whole piece.

(As an aside to the matter at hand, it was interesting to see the effect of a quick link versus a post on kottke.org. On January 31, 2013, Jason linked to a post I did on Sesame Street tweeting a 21st century version of The Monster at the End of This Book. In the week following his link, 781 visitors came to the candler blog through kottke.org. Since Jason posted the current piece to quick links, 1,173 visitors came from kottke.orga and another 105 came through Jason’s tweet. Either Jason’s influence has increased in the past two years or visitors to kottke.org are more likely to click on a quick link than a hyperlink in a paragraph. Or, most likely, both are true.)

What I’ve Learned

For me, there’s little difference between publishing my own site and publishing on Medium. While the fact that the 1,240 views on Medium came to the piece without any promotion on my part is impressive (and something that never happened to me in the days before posting links to social media and notifying other interested sites), it’s not enough to make me want to publish most future writings to Medium. Still, I’m smart enough to know I’m not a typical web writer in 2015.

When I started the candler blog in 2009, the best way to get your thoughts (especially longer thoughts) out onto the web was still to start a blog. I’m lucky to have this site and a readership that still drops by even if I don’t publish anything for weeks on end. The allure of Medium is that you don’t need any infrastructure whatsoever; all you need is your name and your words. For many that will likely be the way forward.

Getting the candler blog to be what it is today was a years-long journey, and it is by no means complete. Personally, I’d rather own every last bit of my writing and how it appears online. Perhaps that’s an outdated thought. But at least now I know: I can do just as well for myself on my own as I can on Medium. If not better.

A Final Thought

One last part of this exercise that bears mentioning. At the very bottom of my Medium post, I left a link to the original article on the candler blog. It didn’t result in even a single click. While there is little incentive to follow a link like that, it does make me wonder whether or not Medium is an island.

When you finish reading something on Medium you’re encouraged to move on to the next piece on the site based on your followers and feeds. That feels like trying to make the web smaller. That’s not the web I love. At least not yet.

Further Reading: “Medium Confusion” by me, just over a year ago. My position has somewhat softened, but I noticed after writing the above piece that I’m still grappling with many of the same concerns from December 2013.

(Header image “Writing through the rain” by José Manuel Ríos Valiente on Flickr.)

I’m not done experimenting. You can see this piece on Medium as well.