I couldn’t help but notice the reopening of Italy’s primary blog classification service, BlogBabel. Just over a year ago I wrote about BlogBabel:
“While it is worth keeping in mind that BlogBabel’s ranking is just one measure of the importance of a particular blog, Ludo deserves kudos for the transparency in which BlogBabel’s rankings are calculated.”
Since then, the ranking factors have changed a bit. Currently BlogBabel says the following parameters are considered1:
| BlogBabel Ranking Factor | Description | Weight |
|---|---|---|
| Google PageRank | The “official” global weight Google assigns to a site. (Its worth noting that this is updated only once every 3-4 months and is not what Google uses internally.) | 1 |
| FeedBurner | Number of feed subscribers for blogs. | 0, thus not considered |
| Link/6 | Inbound links from posts on other sites, added within the last 6 months. | 1 |
| Link/1 | Inbound links from posts on other sites, added within the last month. | 0.5 |
| Link | Inbound links from posts on other sites | 0.5 |
The hit parade is a weighted sum of different ranking parameters. Consider a blog which has had 2 inbound links over the last 6 months. The parameter “Link/6″ is weighted at 1 so the blog’s total value for this ranking parameter will be
<No. of inbound links over 6 months> * <weight> 2 * 1 = 2
As with Google’s PageRank, a logarithmic scale is used to relate the value of each parameter for a blog with the maximum value seen for a blog in the BlogBabel system. BlogBabel discusses this with an example, using data from Technorati. The number one blog, Beppe Grillo, has 6921 inbound links, thus a maximum value expressed as log10(6921). The logarithmic value of the number of inbound links for each blog is divided by this max.
| Blog | Calculation | Value |
|---|---|---|
| Beppe Grillo | score = log10(6921) / max | 1 |
| Roundhouse Kicks | score = log10(1883) / max | 0.85 |
| Edit | score = log10(704) / max | 0.74 |
| Macchianera | score = log10(663) / max | 0.73 |
The main parameter used in the BlogBabel top blogs listing is the number of inbound links to a blog as found in the contents of posts. This is done by parsing blog feeds, which contain the meat of a blog, leaving off the blog roll and other “noise”.
What constitutes a link for BlogBabel (to a degree valid for Google and the other search engines)
- Links must be present in blogs indexed by BlogBabel. If someone links to your blog post, but their blog isn’t indexed by BlogBabel, then it is as if the link doesn’t exist. This is true for Google as well, although it is rare to see a quality site not indexed by Google. In most cases where Google doesn’t index a site, it is because the site has incorrectly specified robots.txt settings or has been banned by Google.
- Links must be in a post, as seen in blog feeds. Several interesting considerations arise here:
- Blogroll links are ignored. It is also a fairly safe bet that search engines such as Google, Yahoo! and Microsoft Live Search, heavily discount blog roll links in their ranking algorithms. We do know that Google is interested in Blogroll data, as demonstrated by their Google Social Graph project.
- Not all links in a blog post are in a blog feed. A blog feed may or may not contain the full post. If a blog is configured to show just a teaser, than BlogBabel won’t see links embedded in posts. To check your WordPress feed settings, look at Sepreings -> Reading -> For each article in a feed, show [x] Full text [ ] Summary and in Feedburner, Optimize (2nd Tab) -> Summary Burner. Because of this possible limitation, major search engines will generally rely on page crawl data, rather than feed data.
- The feed must be well formed xml, not simple text (this is true in general).
- The links must be in the same domain format as the blog is known to BlogBabel, i.e. consistent in the use of “www”, such as http://www.mysite.it/ instead of http://mysite.it/. This issue is called domain normalization or canonicalization.
Manual Factors?
Automated search engine algorithms combined with the explosive growth of the web are what led to the demise of web directories, such as the once important Yahoo! Directory. The internet contains just too much data to keep up with otherwise. Yet there is always the possibility that results in search engine are sometimes manually tweaked, known as “a hand job” in the rather colorful sector-specific vernacular. Google has often been accused of some manual adjustments, such as in the case of famous “Google bombing” attempts, i.e. miserable failure. Google usually denies manual intervention. BlogBabel has also suffered the same accusations, although I would argue that citing faulty Alexa data is not a great way to make your case.
What are the right metrics for social conversation?
There is still a lot of discussion in the web analytics community on how to best measure “social conversation” on the web. Or better, how do we demonstrate tangible business value through social conversation? Jeremiah Owyang’s musings on the topic come to mind. As I’ve previously noted, we could count the number of comments a post has attracted. Yet quantity doesn’t mean quality discourse! We could count post frequency – yet again daily noise is not necessary better than the insight a long, well thought out monthly post might provide. In the end, the intent of this post isn’t to judge the appropriateness of BlogBabel’s algorithms, it is just to note they’re not the only ones struggling with this problem.
Illustration 1: BlogBabel Summary for Antezeta.it blog
wikio
BlogBabel isn’t the only game in town, Wikio.it also classifies Italian blogs. They say
The position of a blog in the Wikio ranking depends on the number and weight of the incoming links from other blogs. These links are dynamic, which means that they are backlinks or links found within articles.
Blogrolls are not taken into account and Wikio only counts links from the last 120 days. We thus hope to provide a classification more representative of trends in the blogosphere. Moreover, the weight of a link depends on the linking blog’s position in the Wikio ranking. With our algorithm, the weight of a link from a top blog is greater than that of a link from a blog that is less well ranked.2
Thus their algorithm is interested in the number of incoming links over the last four months and each link is weighted based on the number of incoming links to the source blog with the outgoing links. As with BlogBabel, Blog Roll links are irrelevant.
The moral of the story
Blog rankings depend on writing interesting material often which key movers and shakers will then link to. (If this post was useful, one could link to it, no? 
)
Similar Posts:
- Blog statistics with BlogBabel at ZenaCamp in Genoa, Italy
- SEO Session: Up Close With Google Blog Search
- Social media measurement and an example, this SEO Blog
- Remove WordPress version information from your blog and feeds
- Web Analytics Optimism in Italy
Registration is now open for the next SEO Course (May 14 and 15) and Google Analytics Course (May 9 and 10) in Milan. Don’t miss the opportunity!
Sean Carlos is a web marketing consultant & teacher, assisting companies with their Search (SEO + PPC = SEM), Social Media & Digital Media Measurement strategies. Sean first worked with text indexing in 1990 in a project for the Los Angeles County Museum of Art. Since then he worked for Hewlett-Packard Consulting and later as IT Manager of a real estate website before founding Antezeta in 2006. Sean is an 
Charts’ weights are inaccurate. PageRank seems to be not considered in new release.