04 August 2009 | Author: R. Falconer SEO Consultant

Google’s Vince update, a different perspective

Google's recent update "Vince" has seen many sites with a strong brand presence see big ranking increases on high traffic, non-brand terms. The update was noticed in US results at the start of the year and went live on UK results a few weeks ago. There's been a lot of speculation since but are some SEOs barking up the wrong tree?

Vince has been great news for some brands. Many have leaped to top positions in Google for some of their highest traffic driving keyphrases. Sometimes brands are appearing for terms that do not appear anywhere on the page or even in the backlink anchor text. So what is going on?

As yet no-one in the SEO community appears to have a comfortable answer - most seem to be going down the road of listing factors that could indicate a big brand. It's possible that this is because most are asking the wrong question, i.e. "How does Google decide what is a relevant brand to push up the rankings?" A better question might be "What has Google done to improve relevance in results for certain terms that has lead to an increase in position for certain brands?"

It has been suggested that Google has introduced a new algorithm that takes into account factors such as the number of mentions in the news, Wikipedia pages, direct traffic to the site, linking patterns etc. but this seems unlikely and unnecessarily complicated. Logistics and feasibility aside, it may or may not be possible to put together a clunky brand algorithm that does something similar to "Vince" using this sort of data, but the actual system appears to be too effective to be doing things this way.

The fact that only certain queries see the Vince effect also has to be taken into consideration - how does Google figure out which query terms to "Vince up"? Introducing yet another algorithm is possible but based on what and for what purpose?

Here is a much simpler theory:

Google's aim, rather than simply to boost brands is, as usual, to increase the relevance of search results. Helping sites with strong brands is the result of rather than the purpose to the Vince update. The sites that are getting a leg up the results are not benefitting purely by having the strongest brand but rather by being the most relevant brands to the user intent of a given query. What it seems that Google has done is find a very efficient way of placing the sites that users want or expect to see into their search results, effectively pre-empting the user's next search.

How it might work

Google (along with the other major search engines) displays "related searches" along with search results. Often, brand terms can be found within the related searches suggestions.


Google's related searches

Microsoft published a patent application which looks at a number of ways in which "related queries" could be calculated. Bill Slawski does an excellent examination of it here. In the patent, it describes how a query log session algorithm could be used to look for search terms that have been used within the same search session. It's easy to see how this could help identify relevant sites to show for certain generic terms.

For example, if lots of users search for "spoons" and in the same session they also search for "Acme Cutlery", it's a safe bet that boosting the Acme Cutlery web site for the query "spoons" will help certain users find the site they want more quickly, providing a quick relevance boost to Google's results.

Query analysis such as this could identify in a single search session, patterns such as:

1. query - "spoons",

2. possibly click on certain results

3. query - "acme cutlery"

4. click on Acme Cutlery site

If this pattern occurs in a high enough percentage of searches, Acme Cutlery gets moved to the first page of the results for "spoons".

Using this relatively simple system, only certain queries would be "Vinced" and only the most relevant brands would be boosted - not just any old brand who happened to be in the news a lot. It also covers the fact that it's not only big brands with high levels of traffic and high news profiles that have been affected. Some smaller sites have also seen big jumps in their rankings on terms that are related to their brand.

The system described above is the simplest one (that I can think of) that could produce the results witnessed since Vince update. It avoids the problems that introducing new algorithms would cause and neatly fits into Matt Cutts' description of Vince too.

It's impossible to say for sure that this is how Google is achieving the end result. Analysing query logs does appear to make more sense than building insanely complicated algorithms though. It's also likely to be harder to spam.

Home | Careers | RSS | Contact Us | Newsletter

International sites: