I have merged the internet population data from the CIA FactBook (reliable source but not the most updated one) to a an (unofficial) Alexa toolbar users distribution and this is what I got:
|Country||Alexa||CIA Fact Book|
Alexa collects data with its toolbar – Did you install it? (I didn’t…) which mainly have implication in regional bias as the distribution of Alexa toolbars is not proportional to the number of users in different countries and as it seems Alexa doesn’t do any “normalization” of the data to fix this bias and doesn’t have any official report on their toolbar distribution – Alexa on Alexa is unavailable (???)
“Alexa Reach” is relative – the percentage of all Internet users who visit a given site rather than absolute number of users which is causing unexplained changes since the “total- distribution-pie” is changing all the time…
Alexa tracks “daily unique visitors” (a.k.a. “Alexa Reach”) and “daily page views” which isn’t like tracking unique (session-based) visits and I think that how many times did you refresh your facebook home page today (could range from 0 to 500 a day ;-) isn’t as interesting as the fact you did visit it today.So…
I don’t have a better alternative to measure other-people-sites-traffic yet (…) but I think Alexa can have much more interesting data by simply moving to more interesting counters (e.g. unique visitor), publish their toolbar distribution (a.k.a. Alexa on Alexa) and start doing some “normalization” with internet population data.