Question Metrics

For anyone interested in that sort of thing:

From the data of the last 90 days:
Unique visitors to the site: 20,755
Unique sessions: 187,120
Pageviews: 756,443 (which projects to a yearly page view total over 3 million), 522,152 unique views (meaning that page had never been loaded by that visitor before).

Visitors from countries
USA: 62%
UK: 10%
Canada: 8%
Australia: 5%
Germany: 2.5%
and the remaining top ten (in order of most visitors): Netherlands, New Zealand, Finland, Sweden, and France.

Given that the stories are in English, this is not entirely unexpected.

Regular visitors to this site
People who visit once per day: 969
People who visit once per week: 2642
People who visit once per 14 days: 4040
People who visit once per month: 7569

90.1% of all visitors are returning, 9.9% are new to the site.

51% of visitors are using a desktop computer
30% a mobile phone
19% use a tablet

(that's a big change from the numbers of the early days of the site... possibly a reflection of changes made to make the site more responsive for small screens?)

How to people come to the site?
50% come here directly, using our URL
22% come here via a search engine, predominantly Google with a few from Bing.
15.5% come via social networking - the Wikia mostly. But also some from deviantArt. A small number also come from Netvibes (which I've never heard of), Pocket (same), Twitter, reddit, and Facebook.
12% come from referral - mostly Bigcloset and TVTropes. A couple web comics are on that list, so I'll give them some credit to reciprocate - veritycomic.com and grrlpowercomic.com (the latter I read, of course, but I hadn't heard of the former... turns out to belong to someone who has written some IF stories for Whateley). Show the referral sites some love for sending people our way.

Analytics doesn't distinguish. It just says UK

Our visitor numbers right now are the highest they've ever been. Especially compared to last fall, when we were seeing our lowest numbers and I was posting to ask what people thought we could do as incentives to bring in visitors or change how we monetize the site to pay for itself.

For example, the numbers for Oct-Dec 2016 (same number of days as the sample above) were

116,006 total sessions
15,542 visitors
437,246 pageviews

every day visitors totaled only 661
at least 1 visit per week was at 2036
at least 1 visit per 14 days was 3530
and at least 1 visit per month was 6064

---

And for the identical quarter (Apr-May-June) from last year:
138,105 sessions
18,028 visitors
536,336 pageviews

So, what we lost from last spring to last fall, we've not only recovered, but added that much more on top of it.

As for explanation?

Well, some of the increase is a result of increasing our content. More new stories would mean more pages to hit. In theory, if a visitor only comes in to the front of the site and reads just the new content... that's 2 page views. And if we boost to a second content on the same day, it could be as little as just 1 more page view... ie, loading the front page and then pulling up both new stories from their sidebar link in other tabs. Adding the third content on a second day could add a second visit to the front page of the site, bringing the total up to 5 total hits. So going on a minimum views: 1 content produces 2 views. 2 content only 3 views. 3 content on split days is 5 views.

So, some of this increase (in pageviews) is simply a result of a boost in content and releasing on two days instead of 1. BUT, that doesn't account for the rise in user numbers... especially the unique users making regular visits. And you can see some of this boost via the increased activity of the forums which isn't as easily tracked with Analytics without me setting up some specific flags or something in the forums, but that I can see on the site itself.

Which is ironic given that unregistered can't post. That's a lot of lurking in progress.

The story views aren't really trackable that way... because Analytics is looking at the whole URL in evaluating views. The URL... not the database entry. The site hits are more accurate to some degree because those are tracking the number of times the story is loaded for a page build. But even there, there is some 'loss' of numbers. For example, most of the front page news gets only a couple dozen hits. That's because even those the news is being loaded its not being loaded as an individual item, its loading as a list of header files from the front page news category. The fact that the entries are all short enough that they aren't cut off before the end of the header is beside the point. It's only when you actually open the full news article by clicking on its title that you trigger a page count.

So clicking on the 'Stories' menu button doesn't count as a page hit for any of those stories. Nor, I believe, does clicking on the Readmore link. BUT, clicking on the title and loading the full story does. As does going to any of the collections and clicking the title of the story in those tables.

Realistically, the only 100% accurate system would be for people to click a vote or 'like' / kudos button for any story they read. But lets face it, MOST people aren't even logging in to use the site... which limits how much we can use those kinds of things. The only thing I CAN do is make sure people aren't hitting stories multiple times to add 'extra' views... which is why if you load the same story a number of times in a row, it doesn't add to it every time.

And keep in mind that story counts are only accurate for the operation length of the site. Meaning, any count total that you see is only the number of hits since Oct 24 of 2015.