I’ve been playing around with some visitor statistics gathering programs over the last few days. For years now, I’ve had some standard tools provided by my ISP, and I’ve written some simple scripts myself. These give me a pretty good overview, but sometimes it’s a little hard to see the forest for the trees in all the data they provide, and it’s always a bit of an issue to distinguish the ‘real readers’ from all the robots and spammers on the Internet.
In particular, I’ve just installed a program called Piwik, which inserts a Java applet into the blog pages for statistics gathering. Since for the most part Java applets are only executed by real web browsers, I now have a better idea of who my real readers are and what they’re looking at.
Piwik doesn’t detect all my readers, for example it doesn’t necessary know about people that use a blog reader (RSS), block Java script, use certain kinds of ad blocking software or use nonstandard browsers. There is however the suggestion that most of the readers it does detect are actually taking the time to load pages into their browser and look at them. My faithful readers, if you like.
One of the real revelations was where my readers are located. Have a look at this list of readers, just from yesterday, but also pretty representative of the last several days. The numbers indicate distinct readers in each country.
United States: 199
Great Britain: 22
Unknown: 8
Canada: 7
United Kingdom: 6
Australia: 5
New Zealand: 3
Argentina: 2
Belgium: 2
Indonesia: 2
Denmark: 2
Bulgaria: 2
Islamic Republic of Iran: 2
India: 2
Netherlands: 2
France: 1
Greece: 1
Croatia: 1
Germany: 1
Czech Republic: 1
Mexico: 1
Guatemala: 1
Spain: 1
Brazil: 1
Tuvalu: 1
Hungary: 1
Ireland: 1
Thailand: 1
Taiwan: 1
Venezuela: 1
Portugal: 1
Japan: 1
Wow! What a diversity of countries and cultures. It’s a little strange the UK is separated from Great Britain, not an important distinction in my opinion. Perhaps if you live in Northern Ireland you won’t agree?
On other days for example, I’ve seen a lot more readers in Brazil, France and Belgium. I’ve also seen a few Afghani, Estonian, Egyptian, Israeli, Swiss, South African, Peruvian, Romanian, Polish, Norwegian, Swedish, Korean, Slovakian, Austrian, Italian, Saudi, UAE, Singaporean, Finish, Belize, Malay, Hong Kong, Serbian, Chinese, Russian and Ukrainian readers.
Since most of these people are readers of garden blogs in general, it’s probably a lot more representative of our community as a whole. What an interesting group of people we are!
I’d love to hear from all of you, and find out who you are and what you think of this blog! I hope you all leave comments sometime. You’re welcome to do this in your native languages if your English isn’t good.
Canada here! 🙂
You might want to look into using something like google analytics. Very comprehensive and very complete. I look at every one of your posts and normally am in Italy, but read through the rss feed so I probably wasn’t counted. Google analytics approximates the readers through rss feeds, I think…
Hello Canada! 🙂
Hi Keith,
Thanks for the suggestion! Yes, indeed, piwik is intended to be an open source alternative to Google analytics, and works in a very similar way. I think it also gives similar statistics.
While I like Google, and use the search engine regularly, I’m reluctant to become dependent on them for something like this. Also, this sort of data about my website is a little sensitive, and if I use Google analytics I have no control over how Google themselves may decide to use it.
Google does not have a very good track record on privacy issues, and there is often talk about what will happen when Google becomes ‘bad’. By this I mean Google now seems to support open source, and gives their services away for free, but both those things may end some day.
For the RSS feed I am currently trying a WordPress plugin called Feed Statistics:
http://www.chrisfinke.com/wordpress/plugins/feed-statistics/
I think this is intended to be a public domain version of Feedburner.
Currently, on the front page of this blog, on the lower part of the right sidebar, is the present number of feed subscribers. This is probably the most important statistic it offers.
Feed Statistics puts a ‘zero pixel image’ (invisible picture) at the bottom of each page, the intention being that most RSS readers don’t display the entire post initially, so anyone clicking on a particular post and loading it in it’s entirety can be counted as a ‘reader’ of that post.
Feed Statistics also puts indirect links into the posts, so if you read a post and click on a link, it reports that back to my site and gets recorded as a ‘click-through’.
Feed Statistics doesn’t offer nearly as much information as piwik, and for example doesn’t give me any information about any readers, where they are located or what they read.
The only thing Feed Statistics reports is total numbers, like x number of click-throughs on this link or y number of people read that post. The statistics it gives are really very vague and not terribly useful. They are also different from the statistics I get from piwik, so it’s not really possible to combine them into something coherent and useful.
If anyone reading this blog has any questions at any time about the sorts of tools like this I use, please let me know. If you have any concerns about your privacy, or you would like me to explain how you can block my recording of this information about your viewing this site, I’m only too happy to help out and explain things.
It’s my intention to be totally transparent about this sort of thing, and respond seriously to concerns anyone has.
Nothing I am currently using can be used to track what an individual person reads, nor am I at all interested in that sort of information. I take very seriously the privacy of individual users. All these tools do is give general information like the number of readers from each country, or the number of people reading a particular post on a particular day. It’s only so I know in general what people like or don’t like about this blog, and what gets people’s attention.
For the most part, these tools are smart enough to only count a single person once. In other words, if you come back and read the same post 10 times, generally it will only count this as a single reading. This may not always be true if you return at a later time on a different day.