Thanks for sharing Rachel!
My objective with all of this analysis is to try and determine if the statistics on Medium can be used to gauge the quality of ones own write or articles.
Your point about sending traffic here from outside sources, traffic that doesn’t have the ability to “recommend” a post, throws the data off kilter a bit. I’m not on Twitter or Facebook, nor do I have an outside blog, so all of my traffic is internal to Medium.
Trying to compare follower to view, to read to rec ratios is a bit complex too. Since it seems that people who have been here a long time appear to have a lot of followers that aren’t active anymore. So that throws a wrench in that data set.
I put your data into the spreadsheet and added an average line, which is the red one at the bottom:
My averages for your comparison:
47.7, 27.8, 62%, 9.5, 27%, 44%
Analyzing mine and your is a bit different, because of the variance of internal and external traffic.
One of the confounding aspects, is trying to determine what the numbers really represent. Since there are so many hidden variables.
If I try to use the numbers to determine if a particular story of mine is better or worse then my other stories or other peoples stories, well, I don’t know that it is really possible to do so. But I’ll continue to try and do so for a while.
As you noted regarding interest in a particular post or story, if it is subject driven, then the readership and/or recommend ratio, etc. would be in part determined by the readers interest or lack there of, in the subject.
A lot of my stories are comedies, so the subject is somewhat irrelevant, and the style of the comedy likely has a greater like/hate influence. Thus, presumably I can compare the data from 4 comedy stories that I write to see which one people liked the best and which one they disliked the most and then study what the differences were between them to analyze my writing technique and story telling.
For you, it strikes me that you would need to compare stories that were about the same subject, say if you wrote 4 stories about trebuchets they could be compared directly.
A further complication is what exactly views and reads represent. I had presumed that views were based on people seeing a story in their feed, but after reading the help article on it, it appears that they are based on people actively engaging with the story. Opening it up. And the read number is based on the calculated time required to read it and how long someone had the story open, not if they actually scrolled all the way down it or not, so that brings about further complications in ones analysis.
This may be a case where the more we know, the less we understand.
As someone else pointed out to me, the way in which Medium improves ones writing the most, is via honest criticism from other writers who’s work one respects. I’m shifting more towards that belief the more I dig into the statistics here.