Technorati Alters Inbound Link Calculations

Posted on September 30, 2005

Technorati has changed the way it ranks inbound links from blogs. Instead of ranking blogs based on inbound links from just the homepage of blogs they will now rank blogs based on the total number of inbound links from blogs from the last six months.

For URL search, we've been looking closely at how we calculate the number of links and sources pointing to a blog, and we've made some tweaks to the display to better surface recent blog activity. Technorati now displays the total number of links from blogs over the last 6 months. Up until now, we displayed a count of all links from blog homepages, which tended to weight more highly blogs that have been around for a long time, even if they have not been posting recently.

The company says the change affects how it ranks its over 18.5 million blogs. Multiple links from the same blog will still count as just one source.

Technorati determines a blog's ranking based on the number of links from unique blogs over the last 6 months. If John links to Susie 5 times in 5 months Technorati will count 5 new links to Susie from 1 unique source. Susie's rank will change based on this 1 new source.

Technorati says the new ranking system will allow blogs that have been more popular in the last six months to rank higher. Technorati gives Interdictor as an example of a blog that has been popular recently and now ranks higher using Technorati's new inbound link measurement. Interdictor received a lot of inbound links for its coverage of Hurricane Katrina and it now ranks #90 on Technorati's 100.

Corante's Get Real offers an innovative idea of what Technorati should do next -- let users query the ranking data in Technorati.

What is still missing? User selection of the period of time used, and a way to select specific areas of authority. The typical query would be -- if Technorati would support this -- "show me the top 100 bloggers on the topic of "social media" based on their posts in the past 12 months". We need to open the model so that user preferences drive the searches, not some canned algorithm. And since Technorati has gone so far with tags, why not use those to determine topic?