Google traditionally uses the date of creation or modification of documents when ranking search results. For instance, the date of creation tells Google the pace of link growth and how soon domains change server location and registrar accounts. These indicators are used to detect signs of spamming. Google also looks at the inception date of a document when processing search queries about breaking news in order to give priority to newly written blog posts and news articles.
As Google simply puts it, old and newly created documents "may be more favorable" than the other. The question is when.
Top Results "Posted 2 Hours Ago"
Google looks at the recent rate of changes in the frequency of query terms to determine what topics are "hot". When query terms immediately reach a certain volume within a relatively short period of time, Google assumes that such searches pertain to a recent news or event. Google will then adjust the SERP score of cached documents to increase the ranking of newly created posts. Thus, old pages that used to occupy the top 3 SERP spots are likely to be replaced by newly created pages even if they have lower link popularity than older ones.
There are more signals used by Google in determining time-bound searches:
- A sudden, remarkable change in the number of search results returned for a particular query phrase or terms - As more people suddenly show interest on a particular topic, more websites would write about it, resulting into steep increase in the number of search results.
- Sudden increase in CTR of recently created posts - As more online surfers click newly written pages (even if they rank below older documents), Google gives more points to fresh content. But how does Google weigh old and new documents when processing queries that are not time bound, like "buy books online"?
When Old Is Better
Through its "search by date" filter, Google can know when to give SERP advantage to stale documents. Old or stale documents that won't rank otherwise in top pages can suddenly dominate the SERP when researchers use "1980-1990" date range, for instance.
But even in the absence of user-defined date filters, Google has a method of determining when to give older or rarely updated documents more points. In my opinion, this could be based on factors like relatively high click-through rate of old documents or higher frequency of certain query terms in distant past.
How Google Defines The Ideal Date
So what's the standard date of creation or update used by Google? It's the average date of creation or update of all the documents in the search results. Thus, documents that are not updated for a certain period of time "may be more favorable" than recently updated documents depending on the "average date of change" of all the search results.
How Old Should Your Homepage Be
Unlike in the case of blogs or news sites, changing the content of the homepage of static websites like corporate sites requires more caution. It's better to google first the target keywords and check if they return newly created documents, news or blog pages. A lot of queries will return a combination of old and new documents in top 3 pages, suggesting that the age of a document perhaps is not a major ranking factor. But if the queries rarely return recent blog posts or news, then it is safer to seldom update the homepage. Otherwise, consider adding news feeds, teasers and feature articles on the homepage.
Being "fresh" is not strictly an on-page endeavor. To optimize a stale or old document regardless of date metrics, use off-page strategies. According to Google, the date of creation or update of documents linking to your page reflects how updated or stale that page is. So even if the content of a homepage is seldom updated, it can still rank among blogs or fresh content if it has backlinks from frequently updated sites.
Other Google Research:
2 thoughts on “When Do New and Old Pages Outrank The Other?”
We always hear that Google love fresh content, but that will not rule our relevancy and user experience once they visit that web page. Very insightful article, one that I seldom see people talk about. Thanks for sharing.
Thanks for an interesting post. One point that you are not mentioning here is the old google PageRank algorithm and its many simulators. The idea behind these is the more times Google revisited your page the more confident it is about its content. So the conventional thinking was that an older page would outrank a page that has not been around for a while since Google is still not confident about its content. However, if you look at links as you say and in particular social sharing a page without any history could be outranking one that has been around for 10 years but this could only be temporary as long as the social sharing process is active. For example a couple of months after a blog post was forgotten and no longer attractive to social sharing or as identified in the “Mashable velocity graph” we could assume that Google will also downgrade a page? Anyway, it would be good to see some studies about a life time of a page on SERPS.
Comments are closed.