What does it do?
Tests how popular this website is compared to other websites, what countries traffic is coming from, and whether popularity is rising or falling.
Why is it important?
This is a simple at-a-glance performance measurement of how popular a website is versus others. It is particularly useful where more detailed information is not available (e.g. most organizations don’t have access to their competitors analytics).
How is it measured?
The Alexa ranking is used. This is based on a very large sample (several million) of web users, all of whom share many browser toolbars which record their viewing history. Generally Alexa can give rankings for websites in the top 5 million – results are much more accurate the higher the ranking. There are some biases in the data, but generally it is still very useful as long as it is used appropriately in context (e.g. more as a relative measure, not an absolute – “website A is more popular than website B”).
Alexa can only measure the popularity of entire domains, e.g. www.example.com. Sites with paths, e.g. www.example.com/news cannot be reliably tested. In these cases, Sitebeam awards 75% of the score from the whole domain to the partial site as an approximate. These occasions are clearly labeled on the results.
The Homepage URL is used alone to determine traffic rank. As it is not possible to merge multiple domains from a rank, subsequent domains are ignored.
If the Homepage URL contains a path, then whatever score is awarded is reduced by 25%, and the site is flagged as belonging to a larger domain. Alexa only holds information for entire domains, and therefore it is not possible to get accurate traffic rankings for paths within a site.
The change in popularity over the past 3 months has an impact on the final score; sites which are gaining in popularity score higher than those which are losing popularity. 60% of the final score comes from absolute popularity, and 40% from the change in popularity.
In the event Alexa does not return a result, the test reports an Informational score.
The use of Alexa is contested by some web designers. We advocate its use, as long as it is accepted that it is not a perfect measure, but merely a useful comparison between websites(particularly where no alternatives are available).
Where more accurate alternatives are available (e.g. if you use Analytics on your own web properties) they should be used instead.
Only whole domains can be reliably tested (see Technical explanation). If a site is filtered by path, the score is reduced by 25% and becomes less accurate. It is not possible to determine a more accurate score given the limits of the data available in these instances.
There are some biases which are identified by Alexa themselves:
- Users are disproportionately likely to visit sites that are featured on alexa.com such as amazon.com and archive.org.
- Alexa’s sample includes users of Internet Explorer, Firefox and Mozilla browsers. The AOL/Netscape and Opera browsers are not supported, which means sites operated by these companies may be undercounted.
- Accuracy in different parts of the world may vary widely due to advertising locality, language, and other geographic and cultural factors.
- The Alexa Toolbar turns itself off on secure pages (https:). Sites with secure page views will be under-represented by Alexa.
How to improve this score
The only way to improve this score is to be gradually improving the popularity of your website. Don’t assume that your websites must always score a perfect 10 for this test. See Perfectionist fallacy.
See also Improving SEO.
How to use this test effectively
This test is useful as a quick comparison between websites, and as a trend to monitor (it is going up or down each month).