20 thoughts on “Numbers

  1. Eh, would probably get too competitive. Like the U.S. News ranking for colleges, or something. I bet they’re trying to avoid that, or trying to avoid how pathetically dismal they are…

  2. I’ve always thought this would be interesting too. Of course, print magazines can prove their circulation based on how many copies they print, sell, etc. But online journals? An editor could simply say “five thousand hits” and we’d have to believe them … or is there a way to prove the traffic beyond simple trust?

  3. Hmm, it didn’t occur to me that people would lie, but I suppose they could if they wanted. It would just be an interesting component to the decision making process about where to submit. Also, I wonder if there would be any surprises. Sites that seem inconspicuous, but which have large readerships. Or sites which seem more popular than they are, if that makes sense. Of course, blogs that sell advertising space are pretty open about their numbers, since that’s kind of how they justify the price of ad space.

    As for Susana’s comment about competitiveness: shouldn’t journals be competitive about how many readers they can bring your work?

  4. Many long standing print magazines include their circulation statistics in their issues (either on the Masthead page or near the back). I think it’s valuable for writers to be aware of the readership of publications where they submit work.

    CLMP keeps track (to some extent) of online mag stats, but I wouldn’t trust those numbers based on what they report for No Tell Motel. I have no idea where they got those numbers — perhaps they asked me years ago when the magazine was new and I have now forgotten giving it to them

    http://timothy-green.org/blog/2010/02/horn-tootin-online/

    In 2009 NTM unique visitors ranged from 4000-6500 per month. This doesn’t include people who soley read us via RSS feeds.

    1. Thank you for sharing these stats, Reb. And for sharing that link! I see that someone very recently has taken a thoughtful look at the same issue about which I couldn’t be bothered to offer more than one line. Huh. I’ll definitely give that a read.

  5. I think web stats can be intensely useless for determining readership. Unlike a print journal–which no one accidentally picks up because they were searching for something else in the bookstore–people arrive at internet sites for all kinds of reasons, and aren’t necessarily readers. The longer the site’s been around, the bigger that margin of non-reading hits is going to be, just because there’s more content for engines to index.

    That said, print journals have similarly obscured circulation numbers. I know one supposedly high-prestige magazine that has just 25-50 real subscribers, and then hundreds of institutional subscribers. So 80% of its issues get shelved directly into other university libraries. It’s always listed as a great place to be published, but I can guarantee you you’re getting read by less than 100 people. The subscription number (500) is wildly inflated compared to actual readers (25-50).

    A better ranking system–and one that would be more complicated to set up and analyze, but would also cover print mags in some ways–would be to look at how much chatter happens when a new issue of a magazine comes out. That would indicate an active readership, as opposed to just a raw number of hits. Which magazine generates the most blog posts/facebook links/tweets/etc. about its content issue after issue? That’s likely to be a place with strong content and good readers, and a place worth publishing in, if exposure is your prime criteria (which asking for traffic stats suggests).

    1. I like this focus on chatter because it seems like it more accurately reflects how people interact w/ content in a Web 2.0 context or whatever.

      But that said — doesn’t the bounce rate number in Google Analytics and also the traffic sources help tell you how many of the hits are actual readers?

      1. Not necessarily. It tells you if they don’t click on any more links within your website. But, if for example, they are linked to one story, and read that one story, and then close the browser—that’s still a bounce. If you’re new content is all on your main index page, like most blogs, then many people will never click an internal link—and are bounces.

        This is the reason why sites break up articles for page-views or only show partial posts on the indexes. It’s just a stupid way to milk hits.

    2. Hi, Matt – I think you’re right, but I don’t think you can even count on subscriber numbers of print journals as an indicator of what actually gets read. I assume that many subscribers don’t read the whole journal, if any of it at all. Or they read all of it, and they pass it to their friends. Or they read none of it, or half a story, or they’re half asleep reading, etc. I also think assuming that institutional subscriptions don’t get read might not necessarily be true – maybe, unfortunately, but I read far more of the journals that come through the English dept – journals I don’t subscribe to, but read when I see something that catches my eye. I read far more criticism and scholarly articles via institutional subscriptions (I’m guessing this isn’t true of creative journals, but who knows.) I think the same is true for online journals, of course, but I think that really all you can get to is a guess about ‘readers’ based on web traffic or print journal distribution. Even measuring chatter – a great idea – I think only tells part of the story, because I think the act of somebody talking about a story or poem in a journal doesn’t necessarily correlate (objectively) to more than a single reader (the blogger) – you could assume that a positive reaction on a blog or website would lead others to read the story, but I think that’s as tentative as assuming that all subscribers to a journal read all of an issue.

      SO, that’s a long winded way of saying that web traffic and subscriptions (both institutional and individual) don’t give you complete info on readership, but I think you can make assumptions based on circulation and traffic as to readership. I don’t think there’s any accurate way to objectively gauge readership numbers, even if you polled every visitor to a website (I might say that I’ve read Man Without Qualities, but have I?)

      I’d ask why does anybody care? I think I care that people read my writing, but why does it matter how many? I’m not saying that I wouldn’t like to have a lot of readers, but I do wonder why it matters. If a billion people read Twilight, does that make it literature? If only a fraction of that have read Don Quixote, or Finnegans Wakedoes that make it a lesser text than Twilight?

  6. I agree that there are other indicators of readership–but I’d argue that you wouldn’t want to completely disregard the raw numbers, flaws or not. Things like search engine indexing can be factored in to their interpretation, after all. Part of the reason I’d want to see the numbers in the first place would be to see if there were any “silent” readerships–maybe elimae is read by everyone in Australia, for instance, but they don’t participate at HTMLGIANT, so they wouldn’t show up on your Active Chatter Index. Or, on the other side of the spectrum, to see whether there are journals who get more buzz than traffic.

    This began with my catching myself in the midst of a line I’ve been telling myself and others for a while–that online magazines are simply a superior vehicle than print, if simple exposure is the criterion. In theory, yes, online has no upper limit on readership, and the work can more easily be shared, linked to, etc. But is it? Is it shared? Are these journals stormed by readers? Or is the truth that for all the medium’s potential, their readerships are comparable to that of all the boring print mags (Blank Quarterly, etc.) that we’re fond of railing against? I realized it would be interesting to know.

    1. Yes, this is interesting also because there are still a number of people who read content but are leery of participating in public forums. The htmlgiant people are always talking abt the private emails they get from folks who don’t want to post their responses publicly.

  7. Web traffic is not a great indicator of readership but we use Google Analytics to track traffic. We average about 6400 unique visitors a month and 80-85,000 page views. The real information you want to focus on, though, is bounce rates and how long people are staying on the site. Unique visitors, site visits and page views include people who randomly happen upon your site via searches, etc. and as such you’re not getting an accurate sense of how your magazine is being read. You have to look a little deeper. We’re lucky to have a really low bounce rate (.98%) so we know people are coming to the site and staying. We know what they’re looking at, how long they’re reading and that kind of analysis is what we focus on more than the number of visitors.

    1. This is fascinating. Thanks for sharing it. Do you ever use this information to change elements of how or when you present your content?

      1. We certainly used the data to update our design such that it would encourage readers to stay longer. It also reminds us to keep the blog updated regularly and know which kinds of blog content work best. We don’t use the information as a means of directing magazine content as much though the data does, often, reaffirm that we’re on the right path. Does that make sense?

        1. It totally makes sense. Ditto on the blog content.

          Have you ever had unexpected spikes in traffic, or sudden shifts in visitor behavoir, and tried to determine why?

          1. Absolutely. This month, we’ve spiked to about 13,000 visitors and we’re trying to figure out why. We generally look through link logs to see if PANK has been mentioned on a big site just to understand significant changes.

  8. The Senior Editor at Poetry, Don Share, recently told me that Poetry has spent several years studying their web traffic, in a way I imagine similar to what Roxanne notes for Pank. I don’t know if they have or will make this public though.

Leave a Reply to davis Cancel reply