Sunday, July 13, 2014
Refresh Rates and Traffic Rankings in the Law Prof Blog Network
Blog Emperor Paul Caron has the latest law prof blog traffic rankings up over at TaxProf Blog. If you look over the stats closely, you'll notice that all of the members of his Law Professor Blogs Network are up between 35% and 350% in traffic over the last year, with most of the blogs increasing between 100% and 200%. At the same time, I have noticed my LPBN pages automatically refreshing when I leave the window open.
So I guess I'll lay out my views straightforwardly -- I don't see any real reason to have auto-refresh other than to boost traffic. I suppose that if I wanted to just open up the blog and let the auto-refresh do my work for me, I could be assured of getting the most recent content. But if I leave the window open to a blog, it's often because I am in the midst of working my way through the past blog posts and want to come back to it -- not to have to figure out where I was. It makes viewing a video over time impossible, as well (as Caron himself notes). And if I'm on the page of a particular post, I suppose I might like the refresh to show any new comments -- but that's a pretty niche desire. What's more likely, perhaps, is that a lengthy comment will get "vaporized" by the refresh rates, as this comment thread indicates. (A great post & comment thread, BTW!)
So is the refresh innovation a real improvement in the blogging experience, or just a way to boost traffic?
Are the blog pages forcing refresh? If so, it's odd that Google Analytics would count that as traffic. While it might be cool to show how used you are, if you are actually using analytics to, you know, analyze then refresh traffic shouldn't count.
If the refresh traffic is caused by the browser, blech. My firefox does that with Tabs Mix Plus, and I immediately turned it off. If I want a refresh, I'll ask for one, thank you. I don't need to go to the top of the page, lose a draft, comment, or have an update edit what I'm looking at.
Posted by: Michael Risch | Jul 13, 2014 2:26:10 PM
The new practices on that network don't seem to affect the "visit" counts on Sitemeter, but do dramatically affect the recognition of "page views." So it's interesting that the new blog rankings no longer include visits and report only the page views. To see the difference, look at this screen capture:
Posted by: John Steele | Jul 13, 2014 3:21:48 PM
The refresh rate was one of the things added by the web designer I hired when I acquired 100% of the network in May 2013. The main driver was the addition of Google ads as part of the redesign. The refresh pushes new Google ads onto the blogs and thus increases revenues. We also added Google Analytics at that time, which showed that many readers keep the blogs open in their browsers for long periods of time, so the refresh also pushes new content to those readers as well as new ads. We compared the refresh rates of several sites and spent a lot of time tweaking our refresh rate to try to find the right balance. We also allow our editors to strike the balance differently on their blogs and use longer (or shorter) refresh rates. As Matt notes, the refresh did increase page views (along with many other things the web designer did under the hood to optimize the network's platform).
Posted by: Paul Caron | Jul 13, 2014 6:47:09 PM
The fact that some blogs have auto-refresh and some (e.g., mine) don't makes Paul's precious rankings utterly meaningless. So I'm keeping mine private.
Posted by: Steve Bainbridge | Jul 14, 2014 5:28:35 PM
I'd like to strike the words precious and utterly, which came off as way more snarky than I intended. Sorry!
Posted by: Steve Bainbridge | Jul 14, 2014 6:09:42 PM
I agree with Steve, and I have posted my thoughts here: http://electionlawblog.org/?p=63281.
Posted by: Rick Hasen | Jul 14, 2014 6:24:17 PM
It's against Google adsense policy: https://support.google.com/adsense/answer/48182?hl=en It's also annoying and unprofessional.
Posted by: anon | Jul 14, 2014 7:11:02 PM
It may be annoying and it may against adsense policy, but the claim that it's unprofessional (by an anon comment) seems somewhat under-argued. It is Paul's business, and he has to consider what's good for his bottom line. That said, if the auto-refresh policy is against the google ads policy, which it seems based on the link, then I suppose that's possibly unprofessional inasmuch Paul should keep his agreement with Google, but that is probably a matter for Google to police with Paul. Maybe the issue is that Paul also trumpets (or more accurately, calibrates) the success of blogs within the network as against other law prof blogs. That seems to be part of Rick's and Steve's complaint. But in lots of ways, Paul's "rankings" and his promotion of subject area blogs generally do us all in the academy a service (even when making money off it). So, it might make more sense to think of this dispute as one that occurs between reasonable persons and in good faith. Of course, since Prawfs doesn't engage in auto-refresh, it suggests further that on Paul's rankings (which should have a Roger Maris type asterisk on the LawProf Network ones), we're totally kicking ass ;-)
Posted by: Dan Markel | Jul 14, 2014 11:28:41 PM
FYI: I added my two cents here by way of explanation as to why I am keeping my stats private: http://www.professorbainbridge.com/professorbainbridgecom/2014/07/the-latest-reason-im-not-playing-paul-carons-blog-ranking-game-.html
Posted by: Steve Bainbridge | Jul 15, 2014 1:09:49 AM
Perhaps it makes the most sense for Paul to post two lists. One of the blogs with auto-refresh and one of the blogs without?
Posted by: I. Glenn Cohen | Jul 15, 2014 9:53:17 AM
Or he could go back to posting the visitor rankings, which disappeared without explanation. I don't think the visitor number is impacted by auto-refresh. Going to auto-refresh and removing the visitor ranking (and ranking solely on page views) was a smooth little move, but not slick enough to get past a bunch of academics.
Posted by: BA | Jul 15, 2014 10:03:13 AM
I stopped separately listing visitors last year after bowing to blogger requests to include Google Analytics data in the rankings due to the continued deterioration of the Site Meter service:
Google Analytics does not use the "visits" category that Site Meter uses. When I did the first quarterly rankings after the switch, the most comparable Google Analytics metric, "sessions," produced results that were so vastly different (5x to 10x) from Site Meter "visits" due to their different methodologies that I dropped them from the rankings.
Given the discussion here, I'll bring back the metric in the next quarterly ranking and label it "visits/sessions," depending on the traffic service used by the blog.
I won't respond to the various snarky comments, other than to repeat what I said above: the refresh feature was added for revenue purposes. I've been doing the traffic rankings since 2008 and only added the refresh feature when I incorporated ads into the network in 2013.
Posted by: Paul Caron | Jul 15, 2014 1:53:29 PM
Paul, when we corresponded about this a while ago, you gave me two reasons, not just the revenue reason: "to optimize traffic (and advertising revenues) on the network." You also indicated that the changes were made to emulate what other media companies do.
The problem that so many have noted is that the refresh methodology yields a number that doesn't reflect what's happening and its now being used to generate rankings. I've often seen stats indicating that, beginning August 2013 for LPB blogs, people apparently started sitting on those blogs for 8 straight hours and racking up hundreds of page views in a day. (Below is a screen capture from yesterday, showing an 8-hour session and 71 page views.) If we have reason to believe that that's not actually the behavior that started happening in August 2013 for those LPB blogs -- if that is an artifact of the new methodology -- then perhaps you can clearly footnote that in your future rankings, with asterisks or otherwise. Because as of now you are comparing apples to oranges.
Posted by: John Steele | Jul 15, 2014 2:13:31 PM
Should be "for up to 8 straight hours," as demonstrated in the capture. But I saw 8 hour sessions quite regularly for one blog in particular and you can see from the capture that it's happened as recently as yesterday.
Posted by: John Steele | Jul 15, 2014 2:21:48 PM
In light of the evidence of the 8 hour sessions, etc., it seems to me you can add the auto-refresh "for revenue purposes" OR you can have rankings based on page views which seem to accurately reflect reading of the law professor blogs, but you cannot do both simultaneously.
Posted by: Rick Hasen | Jul 15, 2014 2:33:06 PM
To be fair, this is more typical of the things one started seeing in August 2013 for LPB blogs.
Posted by: John Steele | Jul 15, 2014 2:43:00 PM