Where in The World Does Blog Traffic Come From?

By crawling publicly available visitor data we can get a picture of the distribution of visitors to blogs. In addition to looking at referrals, I've started collecting information about the location of visitors to 1, 500 weblogs (see earlier post for a description of this data). The the tables below show firstly the number of visitors per country ranked by frequency (top 20), and secondly the top 20 ranked by per capita visitor.

Country	visitors
United States	13, 334
Canada	954
United Kingdom	942
Germany	270
France	270
Portugal	213
Spain	166
Netherlands	137
Australia	115
Brazil	113
Italy	100
India	75
Switzerland	70
Sweden	69
Singapore	59
Norway	58
New Zealand	58
Belgium	57
Israel	56
Poland	55

The table below shows the number of visitors per 1, 000 capita for the country. I have removed those countries which had fewer than 10 visitors (which were generally island nations from the Caribbean). These results are very preliminary (being only a single day's collection), but it is interesting to see the ranking of countries beyond the obvious first 2.

Country	visitors per 1000 capita
United States	0.0451
Canada	0.0291
Portugal	0.0202
United Kingdom	0.0156
New Zealand	0.0144
Singapore	0.0133
Norway	0.0126
Ireland	0.0117
United Arab Emirates	0.0105
Switzerland	0.0093
Israel	0.0089
Netherlands	0.0084
Denmark	0.0081
Sweden	0.0077
Finland	0.0069
Australia	0.0057
Hong Kong	0.0057
Belgium	0.0055
France	0.0045
Spain	0.0041

Finally, the image below shows a graphical representation of a portion of the per capita data.

Comments

[Sandro Saitta]

This is a very nice post. It would be interesting to have the same trends in a few years. I'm quite sure that countries such as India and China will have a better ranking. By the way, if you want to know from where do the visitors of your website or weblog come from, you can use Google Analytics. The only drawback is that you can only analyse your personnal site.

[Kamal Nigam]

Your data comes from 1500 blogs with public sitemeter referrer logs (as I understand it). There's some danger here of generalizing geographically without accounting for things like language and sitemeter penetration. For example, If none of your 1500 blogs are written in Korean, you'd expect fewer blog readers from Korea. Or, if sitemeter has minimal penetration among Australian bloggers, you'd expect relatively few Australian blog readers. The numbers and graphics you present are very compelling, but without knowing more about the data, it's hard to know what conclusions to draw from them. A comparison between the geo distributions of the 1500 blogs and their readers might be more interpretable.

[Matthew Hurst]

You are absolutely right, and I've raised these issue in my previous posts. This is purely directional and only after the first collection of data. However, I think that even given those issues this data is worth looking at.