Sunday, February 17, 2008

Friends, traffic and statistical musings

A little while ago Neil had some interesting traffic stats. I wanted to record the figures relevant to this blog so that I did not forget them.

In looking at out-clicks, Neil found that on Old Lines from a Floating Life since April 2006 this blog came in first with 615 out-clicks.

On the new New Lines from a Floating Life, in the period since December 2007 this blog came in second with 94 out-clicks, while on Oz Politics and the Big Archive since July 2007 this blog was 9th with 33 out-clicks.

Now these are quite large numbers. They suggest that perhaps one visitor in thirty to this blog has come via one of Neil's sites.

When we look at the opposite side of the equation, traffic from this blog to Neil's, his stats show a far less positive picture. This blog features in the top source group only on Old Lines from a Floating Life and then with 73 clicks. So Neil has only been getting around one visitor back for perhaps every ten he sends me. I must do better!

My stats packages - Bravenet and Site Meter - on this site are limited, so I cannot do the type of thing that Neil has done. By the way, the difference between the aggregate visitors on the two as shown on the front page is largely due to the inclusion of my own visits in Site Meter.

I have fewer visitors than Neil. Over January, this site averaged just over 53 visits per day excluding my own. Ranked by order, those visitors came from:

  • Search engine references, mainly Google.
  • Then I average 5-7 return visits per day.
  • The rest is referals.

Referals bounce all over the place.

However, when I look at the last 100 visits, I found 8 from Neil's Oz Politics and Big Archive, 2 from Neil's New Lines from a Floating Life, then one each from Catallaxy, Technorati, New England Australia, David's After the Vote and Legal Eagle's The Legal Soapbox. So a spread.

I thought as a matter of interest I might check the detail of searches. After looking at the first forty, I think that the most I can say is that they are very varied!

I don't think that I can conclude much beyond understanding why most people are here for 0 seconds - this blog is unlikely to yield much on Jeff Dunham's personal life or the location of a personal trainer in a Sydney suburb.

I did realise one useful thing, however. The correlation between searches and the most popular entry pages is not strong, simply because quite a few referals are linked to individual stories and hence affect entry patterns.

I also realised that the pattern of clicks from Neil's blogs is affected by his own visits here. More broadly, I can spot my regular visitors. I must look at this some time, because I do have regular visitors that I do not yet know.

Well, I think that I have wasted enough time to today on this!

2 comments:

ninglun said...

WordPress claims not to count my own clicks in its outclick stats, though I tend to doubt that. Certainly it doesn't count my own visits to my own blogs; nor does Sitemeter, since I have trained it not to.

Just for your interest, Jim. Certainly you are quite often close to the top of the 24 hours "top clicks" widget in my side bars.

Jim Belshaw said...

Thank's Neil. That's helpful. For us obsessive bloggers, I fear that we both belong to this class, all this is very interesting in helping us fine tune.

By the way, in response to a comment from Lexcen, I am preparing a post that deals with the challenges faced by Austrlia's demographic future. I am trying to base it just on the evidence, sort of a scoping exercise, although my own views (prejudices?)obviously intrude to some extent. I would be interested in your comments in due course.

I hope to get it up today, although it was not on my priorities. Another itch, so to speak.