tedster - 8:06 pm on Feb 23, 2009 (gmt 0)

Highly visible brands are, well, highly visible. So they naturally tend to be involved in social media conversations, shopping comparison sites, etc. Metrics for determining the effect of any non-linked mentions on regular search rankings are not easy to nail down. It does appear from the coversation mining tools I've used that an increase in mentions (not necessarily linked) is often and quickly followed by imprved search traffic.

The question raised in the opening post is whether something has changed very recently in the way Google treats a brandname - and that raises the question of how "brand" could be defined algorithmically. It's one thing to say "I know is when I see it" and another thing altogether to define the concept of brand rigorously.

I can only say that this is a direction I'm sure Google has been considering for quite a while. They are not running a ranking contest, after all, but a service for their end users. If the end users seem happier with branded results it makes sense to me that they would look harder at this area. Sitelinks were a kind of step in this direction, right?

So is there an uptick in big brands on the keyword searches? How can WE measure something like that when defining the concept of brand, rigorously, is rather slippery thing to do?

I've learned a bit more about those two big brands that escaped the yo-yo this month. In both cases there was a major increase in mentions from various news media, both offline and online. This meant a flood of unlinked mentions, a significat growth in social media mentions, and in one case, more than a half million new backlinks as well. So deciding on cause and effet is not an easy call.