Branding, SEO & The Vince Update
Over the past couple of months there has been a lot of discussion in the SEO world about Google's "Vince" update. Although the change occurred in the second half of January, the discussion in the wider SEO community did not begin until Aaron Wall posted a blog entry on February 25th about Google seemingly placing "heavy emphasis" on branding in their search results (http://www.seobook.com/google-branding). Google, through Matt Cutts, confirmed that an algorithm change did occur. However, Matt stopped short of calling it an update, labeling it instead a "simple change".
Regardless of what you call it, a change did occur in January to Google's algorithm and it did affect the rankings for some queries. But what changed? Does Google prefer branded sites now? And most importantly, is there anything that you should do differently for SEO?
Despite the ranking shifts that were pointed out in Aaron's original post, I don't feel that we should assume Google intentionally increased the rankings of branded sites. I think that the most reasonable explanation is that they made a change that was intended to eliminate the effectiveness of certain spam. At the same time, they may have increased the relevancy value of certain SEO factors that would have influenced big brands as well. The side effect of two such tweaks could very easily be an increase in rankings for branded sites for certain keywords.
Although I spent an entire weekend contemplating the issue, I came back to the office on Monday confident that something more logical (at least in my mind) was in play. Whenever I'm debating a Google algorithm issue, the question I always ask myself is what makes most sense from a search quality standpoint. In the case of branding and its potential effect on rankings, does putting an emphasis on brands improve Google's results? I would argue that it doesn't. While emphasizing certain brands might be more helpful for some searches, it certainly would not make the results better for every potential query out there. And Google will rarely make such an algorithm update unless they are confident that it will improve search results across the board.
Not Update for Brands
I got validation for my initial conclusions while doing research for a segment that I recorded with Susan and Virginia on SEM Synergy (jump to the last six minutes). During my research, I came across a video that Matt Cutts made as part of the GoogleWebmasterHelp YouTube account that was created at the beginning of this year (http://www.youtube.com/user/GoogleWebmasterHelp). In the video, Matt answered a question from a user about the tweak mentioned in Aaron's blog. Watch it now if you haven't already.
As is typical of Matt and everyone at Google, he was not specific in his answer. However, he did say a few things that I think are noteworthy. Here are several quotes that I found particularly interesting (in order):
"We don't really think about brands. We think about words like trust, authority, reputation, PageRank, high-quality . "
"And so the Google philosophy on search results has been the same pretty much forever."
"But it affects a relatively small number of queries . it's not like it affects a ton of long-tail queries."
"I don't think of it as putting more weight on brands. We really don't think about 'brands' in search-quality that much."
"But it's not that we always try to return brands, we try to return whatever we think the best results are for users."
"And so what you should be doing doesn't change."
Conspiracy theorists will say that of course Google is not going to say that they do not think about brands. They wouldn't tell us what they did, right? Although I agree that Google is not going to say what they did, I've also found in my experience that they don't directly lie. They are often vague when they don't want to tell us something, but I've never seen evidence of them outright lying. The quotes above are fairly direct in my opinion, at least when it comes to branding.
Other Possible Explanations
So if Google didn't set out to improve the rankings of branded sites, what did they do exactly? I can think of two relatively simple tweaks that could have been made that would have improved the rankings of established brands:
- They increased the relative value of link age and/or domain age. Well-known brands in general have been online much longer than a lot of other sites. Thus, many of the links going to those sites would be older, increasing their relative value compared to links to newer sites.
- Google may have increased the value of text content that's on the linking page outside of the anchor text. They may have also slightly decreased the value of anchor text. It's also possible that they did both. The end result of any of these three scenarios would be that branded sites that had a lot of links from relevant pages but bad anchor text would have seen a bump.
Both of these tweaks would be pretty common and are quite possible. Another scenario that is slightly more far fetched is that Google has now found a way to associate certain words that appear together often, even if there is no link going to any particular domain. For example, let's consider one of the examples mentioned by Aaron in his post. Radio Shack's Web site started ranking for the phrase "electronics" after this change was made. Interestingly, it now ranks #1 for that keyword.
The two possibilities I mentioned above would have helped Radio Shack. However, by doing some searches in Google you can see that "electronics" and "Radio Shack" have been used an extraordinarily large number of times on the same page. Take a look at the following Google queries:
"radio shack" electronics -site:radioshack.com (≅2.5 million results)
intitle:"radio shack" intitle:electronics -site:radioshack.com (≅4,000 results)
The first query returns all the pages in Google's index on which "radio shack" and electronics both appear on the page. The second query returns all the pages in Google's index that have "radio shack" and the word electronics in their Title tags. The number of results for both of these results is pretty high considering I eliminated pages from the Radio Shack Web site.
The interesting thing that I found when I reviewed a lot of the results from these queries was that many of them did not actually link to www.radioshack.com. This tells me that Google is possibly viewing these terms as related or synonymous in some way and thus the Radio Shack site is getting a bump for that query. I can't think of how it could be programmed algorithmically, so it's hard for me to say that's exactly what's going on. However, that would certainly explain a jump from nowhere in the top 10 to number 1 in the last few months. The ranking factors I mentioned would have had to have been turned up tremendously in terms of relevancy value to account for that type of jump.
Assuming this type of association is what is coming into play, it could explain why Matt would have said that the change "affects a relatively small number of queries . it's not like it affects a ton of long-tail queries." It's much less likely that Radio Shack would appear so often with long-tail keywords. Thus, they wouldn't start ranking for any long-tail terms as a result of the change.
Change in SEO Strategy?
Regardless of what was actually changed by Google, the most important thing to do is determine if the change alters your SEO strategy. I always ask myself the following question when I learn about something Google tweaked: does this change what I do? Nine times out of ten, it does not.
There aren't too many things that will have a significant effect on your site's rankings. Increasing internal relevancy is as simple as tweaking Title tags and non-hyperlinked body content and combining that with optimal internal linking. Once you've finished that process, you'll want to extend upon your strategy by adding inbound links from related third-party sites. When it comes to the SEO for a particular keyword, there's not much else to it. The only thing that changes is how effective each of these things is. But regardless of how much Google discounts these things by tweaking their relative value, these core factors are still the shortest path to optimal rankings.
This recent update does not give me any more reason to recommend branding than I did before. Successfully creating a brand for yourself or your company should always be a goal because it is good marketing practice. However, SEO should never be your main purpose. Keep in mind that SEO benefits would only come from successful branding, not any attempt at branding that you throw out there. In my estimation, the only way branding can possibly affect your site's rankings for a particular query is if your brand becomes essentially synonymous with the search query. So much so that the keyword you are targeting appears on Web pages with your brand name throughout Google's index. But that is a long-term byproduct of successful branding that will happen naturally. You won't be able to make it happen.