Google Once Again Seeks to Add Historcal Data to Ranking Factors
Hold on to your seats folks, Google’s decided to revisit using historical data as part of its ranking factors. This is causing quite a stir for the search engine marketing community. Information retrieval based on historical data isn’t a new concept. Google attempted to introduce historical data to its ranking algorithm back in 2005; however, the USPTO rejected the original patent for reasons that seem to be rather innocuous, according to Bill Slawski of Search Engine Land.
Just like in 2005, these patent applications are guaranteed to cause a stir. Because if these patents go through, Google will be adding another dimension to the way they determine relevancy. Here are just some of the factors mentioned in these patent applications:
- Document inception dates – when the web page was created
- Document content updates/changes – how often is the content updated
- Query analysis – how often people click on the result, and how that query relates to other topics and changes over time.
- Link-based criteria – when do the links appear and leave a site, whether or not the site is gaining link popularity or loosing (which could indicate whether or not a site is fresh or getting stale). The quality of the links, whether or not they’re coming from authority sites.
- Anchor text – how often the anchor text is changed, whether or not it’s staying fresh and keeping up to date with the content on the site.
- Traffic – studying traffic patterns over time to determine seasonality, freshness and quality of the content.
- User behavior – how many times that search result is selected, how much time the user spends on the page, and whether or not the user immediately exits (indicating that wasn’t what the user was looking for)
- Domain-Related Information – legitimacy of the domain, how long it’s been registered and whether or not it’s hosted on a “good” domain server.
- Ranking History – monitoring for sudden spikes in rankings which might indicate a fresh new topic or spamming the search engines. Rate at which users selected that ranking over time.
- User maintained generated data – how often do users bookmark or select the web page as a favorite.
Obviously, it’s a little early to run out and start yelling, “The sky is falling, the sky is falling…” However, internet marketers should make themselves familiar with these potential Google changes. There is some wonderful commentary already available on this interesting new SEO development:
- SEOMOZ.org’s Article on the Google Patent – Information Retrieval Based on Historical Data
- Bill Slawski’s Article, Google’s Historical Data Ranking Factors Revisited, plus Advertising Rankings
On a side note, if you get anything out of this blog post, get this: freshness is becoming more and more important to the search engines. MSN/Live is already weighing the freshness of content in their algorithm, so don’t rest on your laurels…keep your on-page and off-page factors fresh and hold on because SEO is getting more interesting.