HUGO AWARD-WINNING SCIENCE FICTION & FANTASY MAGAZINE
Editor's Desk:
Clarkesworld by the Numbers
I'm a data junkie. Over the last five-plus years, I've collected a lot of data about all things Clarkesworld and some about the state of online fiction. This month, I thought I'd dig into that data and share some of the more interesting data and what they mean to me.
I'm going to start with something that says a lot about the history of Clarkesworld and perhaps even a little about the growing acceptance of online fiction since 2006:
The solid line in this chart indicates unique readers per issue and the dotted line is a cumulative six-month average that influences how I determine our monthly readership.
One of the first things I notice when I look at this graph are the spikes and how they become more pronounced over time. When we first launched in 2006, online fiction was still suffering from an image problem, the audience was much smaller, and there were very few SFWA-qualifying online magazines. None of those problems exist today. The spikes for those early years are small and reflect the lower potential audience of the time. This fits within the context of discussions I've had with other editors that were around at the time.
If the size of a spike can help you guage the changes in potential market, the frequency should demonstrate interest in what the magazine has to offer. It's also a great visual indicator of the viral nature of the medium we're working in. If you like something, you are more likely to share it. If someone shares something with you and you like it, you might just stick around and become a regular reader. In a sense, this is just good old-fashioned word-of-mouth marketing, but on steroids.
I do feel the need to comment here. When we select stories or articles, our goal is to pick great content. As far as we're concerned, spikes happen when people think we met our goal. They are a side-effect, not the objective.
By now, you're probably curious about what triggered some of those spikes, so here's a few:
1. An editorial on the state of short fiction
2. "Spar" by Kij Johnson
3. "The Things" by Peter Watts
4. Hugo and Nebula Award nominations for Clarkesworld, "Spar" and "Non-Zero Probabilities" by N. K. Jemisin.
5. Hugo win for Best Semiprozine.
6. Several items, including a resurgence in activity for "The Things", "Spar", and various non-fiction. Also strong showings for the first part of the serialization of "Silently and Very Fast" by Catherynne M. Valente and "Tying Knots" by Ken Liu.
OUR READERS
We don't know much about our readers, not even the ones that are subscribing to our Kindle or EPUB editions at Amazon and Weightless. (Ok, we do know you're all really classy people who empower us to keep doing what we do!) However, web statistics give us a hint. For example:
70% of our readers are in the USA
20% of our readers are in the UK, Canada or Australia
10% read from a total of over 30 different countries, more than half of which aren't English-speaking countries
35% use an Apple computer
15% use Windows XP
5% use some form of Unix/Linux
Our most widely read story is "The Things" by Peter Watts (240K+ readers)
Our most widely listened to story is also "The Things" (17K+ listeners)
Roughly 5-6K of you listen to our stories on our podcast
According to Facebook:
Obviously, this is more a profile of the 3500 people who are willing to follow us on Facebook (as opposed to the over 25K reading the magazine), but it does paint an interesting picture. This is something I'd love to have better data on, but in the end, it doesn't influence us all that much though it might come in handy if someone decides they want to buy ad space from us.
SHORT STORIES
Since 2009, I've been using an online submission form to collect stories from authors interested in being published in our magazine. These stories make up what is known as the "slush pile" and I occasionally post slush pile statistics over on my blog. Here, I'm going to focus on the big picture.
The solid line is the number of stories submitted to Clarkesworld on a monthly basis. The dotted line is the number of rejection letters that go out in the same month. As you can see, we receive a lot more stories than we could ever publish. The two dips to zero were months we closed to submissions, typically Christmas break. The huge jump after the second reopening maps to a change in submission guidelines (allowing longer stories) and surge in readership and award nominations very close to spike #3 on the chart above.
Let's dig a little deeper:
There isn't a lot to be said about this. I have no way of knowing whether or not what we are seeing here is an accurate representation of the short fiction authors in the science fiction and fantasy field. This data is more interesting in comparison to the following:
Over the years, I've found it very interesting that the gender breakdown for accepted stories at any given period in our history is nearly evenly split, despite the wider disparity between the genders in the volume of submissions. I've seen all sorts of interesting interpretations of this data, but from my standpoint it simply says that when it comes to talent, neither gender has an upper hand. In the end, quantity loses to quality.
Now, let's look at international submissions:
The top eleven are shown, with the Phillipines just missing the cut at number 12. They and ninety other countries represent that "other" wedge of the pie. I'm very pleased by the variety of countries represented, but the top four tend to dominate the stories we publish. In the last year, however, we have had the pleasure of publishing two Chinese translations. This is an aspect of the magazine that I hope to see expand over the next few years and a way data is influencing how I run things.
If you've made it this far, you can consider yourself a fellow data junkie. If you'd like to see more, or have any questions, please feel free to jump into the comments and ask away.
ABOUT THE AUTHOR
Neil Clarke is the editor of Clarkesworld Magazine, Forever Magazine, and Upgraded; owner of Wyrm Publishing; and a four-time Hugo Award Nominee for Best Editor (short form). The innagural edition of his Best Science Fiction of the Year anthology series was published by Night Shade Books in June. He currently lives in NJ with his wife and two children.
WEBSITE
Also by this Author
PURCHASE THIS ISSUE:
CLARKESWORLD ESUBSCRIPTIONS AVAILABLE AT
Follow Us On
ISSN 1937-7843 Clarkesworld Magazine © 2006-2015 Wyrm Publishing. Robot illustration by Serj Iulian.
Bob Jacobs wrote on June 2nd, 2012 at 4:38 pm:
Thanks for publishing the numbers and the supporting information. It's all very interesting, but of particular interest to me are the gender stats because I'm seeing a lot of discussion here in the UK at the moment about gender bias in publishing.
Johann Thorsson wrote on June 5th, 2012 at 8:44 am:
First off: congrats on the increased readership, you certainly deserve it.
These numbers are very interesting, especially the male/female submission ration vs the published stories. Is it possible that men are over-confident in their writing ability and women more cautious?
Neil Clarke wrote on June 5th, 2012 at 10:17 am:
Johann: Anything is possible, but there is insufficient data in my records to back up that hypothesis. We don't know anything about the mindset of the people submitting.
Suzanne Church wrote on June 8th, 2012 at 12:19 pm:
I'm a bit of a numbers junkie myself, so I really enjoyed all of your data charts and summaries.
Clarkesworld is definitely the place to read good fiction and stimulating Non-Fiction.
And the podcasts are very well done.
Congrats!
Tony Baechler wrote on June 9th, 2012 at 3:12 am:
I found your editorial for this month interesting, but less than helpful. I'm totally blind, so the graphs and charts are meaningless to me. It would be nice to get some actual numbers that I can read. In particular, you mentioned the slush pile. I've had thoughts of publishing an SF ezine over the years, but I haven't seriously considered it. It would be nice to know actually how big the slush pile is and how many stories get rejected. I've read on other blogs that the number of submissions is huge, but I haven't seen any actual numbers. Also, how many stories are rejected out of hand and would never be published, such as due to poor writing and such? How many stories could have been published if the budget was not an issue? How many months in advance do you buy submissions?
Neil Clarke wrote on June 9th, 2012 at 3:27 pm:
Hi Tony,
Sorry about that. Given the volume of data I was working with, graphs were the easiest way to show what has been happening. I'll try to answer some of your questions...
The slush pile is currently receiving 600-700 stories per month, with a rejection rate over 99%. Budget is never the reason a story is rejected. I don't like to buy inventory more than six months in advance, so if I reach that point, I'll close to submissions for a couple of months. That hasn't happened yet.
I often blog in more specific detail about the types of stories, reasons for rejection and other details on my blog. I summarized the January 2012 submissions in my last slush update and it includes stuff that may be more interesting to writers and potential editors than this did: http://clarkesworld.livejournal.com/179488.html
As for the charts, a quick overview:
Readership: steadily growing since the start, but not in a steady curve. Spikes in readership at certain issues usually leading to an increase in regular readers. 3k when we launched in 2006 and reaching 30k now with most of the growth in the last three years.
Facebook: Claims that we have an audience there that is 2/3rds male. The largest age bracket is 35-44, followed by 25-34. Ages seem distributed by gender evenly.
Submissions: Were 300-400 before 2010 and spiked to current numbers around the same time our readership started to take off.
Submissions by gender from issues 24-67: 69% from men.
Stories published by gender (all): 51% from women.
Submissions by country from issue 24-67: about 75% from the US, UK and Canada taking about 6% each, Australia with 3% and roughly 100 other countries representing the rest.
Hope that helps.
-Neil
Russel Gaines wrote on October 20th, 2012 at 12:23 pm:
Actually, this does show a gender bias and that disputes your claim of quality over quantity. If women are only submitting 31% of the stories but have 51% of the acceptances, there is clearly something happening. As mentioned by Bob Jacobs, a massive gender bias is happening.
I think this points to editors and not readers and that says a lot.
Neil Clarke wrote on October 20th, 2012 at 1:02 pm:
There is no way you can dispute a statement on quality without having looked at the submissions yourself. Gender bias requires unequal treatment. We don't separate stories into boy-girl piles and review them differently. Most of the time I don't even look at the name of the author of the story I'm reading. We don't have quotas or anything remotely like that. Quality determines the decision.
Quality is subjective. Personally, I'm not looking for stories I've seen a million times before. Same-old, same-old doesn't work for me. Perhaps women are doing less of that, maybe for no other reason than they haven't been a majority voice in our field. Dunno. Statistically, women read more. Maybe that factors in. The point is, we don't have the data necessary to make anything more than the observation that when it comes to quality writing, we're seeing an almost even number of good male and female authors. Hardly a bad thing and it really shouldn't be distressing... or shocking.
I can't speak for Bob, but I believe he was alluding to earlier discussion (elsewhere) of gender bias in UK publishing, which is very male-dominated, particularly in anthologies. Hard to believe some of those editors can't find a single quality story by a woman (or an author to solicit from)... which is where that controversy comes from. That's more in line with what bias is all about.
Erin Hoffman wrote on December 4th, 2012 at 10:39 pm:
Fascinating data, Neil. Thanks for posting it.
I agree with you about virality, and it will be interesting over time to see whether some more identifiable periodicity emerges in your data (seasonality, for instance). With all the spikes it's harder to say. But you definitely seem to carry an advantage from each of them, which seems to suggest the magazine thrives on controversy?