The industry, of course, has done its best to fuel this faith. Through the years we've seen a handful of studies claiming that search really can help drive this brand impact -- from ancient research published by the US IAB to frustratingly incomplete pronouncements from the company that has the most to gain from search marketers bidding on brand terms. Even we at Jupiter have been saying since at least 2004 that you can brand with search; the problem is that we as an industry didn't know then -- and we still don't know now -- how to assign a value to the brand impact provided by search listings. And without a concrete value, marketers have no idea how to adjust their bids to account for branding.

This autumn I'm going to have a deeper look into this issue, and see if we can't start down the road of assigning concrete values to the brand impact of search. But in the meantime, I found a couple of analyses recently that touch on the bread-and-butter of search branding: bidding on brand keywords. Harvest Digital searched on 100 UK brands [PDF] and found that, two months after Google started allowing UK marketers to bid on competitors' brand names, less than 20 percent of brands are being targeted by competitors. (In fact, they found that most of the brands they searched for don't even bid on their own keywords!) Meanwhile, StraightUpSearch argues (with no evidence but great conviction) that there's no point in bidding on competitors' keywords, since the direct response ROI doesn't work out. If that's true, then bidding on competitors' keywords is purely a branding exercise.

As I said, I'll be looking at this in much more detail in the coming months. If you have any insight or case studies you'd like to share, please drop me an e-mail: n e l l i o t t (at) jupiterresearch (dot) com.

Which is pretty odd. Because the search engines and portals have been getting their butts kicked in social media. Sure, Google owns YouTube and Blogger, and Yahoo owns Flickr and del.icio.us. But those were all acquisitions. The truth is, portals and search engines have had no success at all in building their own social media offerings. (And Blogger, Flickr, and del.icio.us aren't exactly hotbeds for marketing. Nor, really, is YouTube.)

So these findings tell me that social marketers are pretty confused -- they still don't know exactly how to make social marketing work, or even where they should be trying. Maybe that's why most European marketers spent less than €10,000 on social marketing last year, and why less than 10% of European social marketers say they're happy with their social marketing ROI.

It's a complicated topic -- one that's hard to deal with properly in a blog post -- but for those interested in learning more, we've just published a new report with lots more discussion and detail: Mining for Fool's Gold: European Social Marketing Budgets Grow Slowly as Marketers Focus on Free Opportunities.

(And for anyone who's curious, Facebook was #2 on the list of most important sites for social marketing. MySpace finished a very distant #4.)

But really, you should follow me on twitter too -- it's great fun, and it allows great conversations.

If anyone will be there and wants to meet up, either stop by my presentation -- Thursday at 3:30pm -- or drop me an email: nelliott *at* jupiter research -dot- com.

But behavioral targeting is incredibly scary to publishers and advertisers and any other company who wants use anonymous user data to effectively target advertising. Because even some very smart people simply don't take the time to understand what's being discussed. I'm sure Sir Berners-Lee is one of the great minds of the digital world -- but either he has never bothered to read up on behavioral targeting, or he doesn't know the meaning of the word 'anonymous,' or he simply doesn't trust his ISP.

It's probably the latter: he doesn't trust his ISP. That's fair; everyone can decide for themselves who to trust. But it's also odd, because he presumably already trusts Google to store all his searches -- including personally identifiable data -- for 18 months. Does Berners-Lee really trust Google more with data about exactly who he is and and what he searched for (which is not just the 'worst-case scenario' he discusses in the BBC story, but the actual fact of his relationship with Google today) more than he trusts BT or Virgin Media with anonymous data? (After all, who stores more sensitive data about you: your search engine, or your ISP? And who has a worse history of leaking users' behavioral profiles: search engines, or ISPs?)

This is why the ISPs who partnered with Phorm just weeks ago have already gone weak in the knees. It seems it's simply not possible to have a fair, informed discussion of behavioral targeting -- not in the face of scary stories from the media, and over-reaching, under-informed comments from the type of digerati who should know better.

The map isn't quite perfect. While there's some data to support the idea that Bebo are the largest network in the UK, there's almost no chance they're bigger than MySpace across all of Europe; our 2007 data has MySpace a clear #1 in Western Europe with Facebook and Bebo fighting it out for #2. Also, Skyblog (listed as #1 in France and #3 in Europe) and LiveJournal (listed as #1 in Russia) aren't social networks, they're blogging sites. (Yes, it's getting harder to tell the difference between those two categories. But even with Skyblog's addition of some networking features last year, these two sites are still, clearly, first and foremost about blogging.) Finally, I'd have thought some of the countries that aren't labeled would've been pretty easy to fill in with a current leader -- among others, it seems clear that StudiVZ is #1 in Germany, and that Facebook is #1 in Sweden.

But despite these flaws, this map is a very useful reminder that social networking is still a relatively fragmented market. English-speaking media types have been completely fixated on Facebook for the last year, forgetting both that MySpace is still bigger globally, and that Facebook until recently had almost no presence outside the English-speaking world. Friendster and Orkut are widely described as also-rans in social networking, despite the fact that they're each the #1 network in a crucial developing market (Asia and Latin America, respectively). And I guarantee that not a single person who looks at that map will have already heard of every network mentioned. (Had you ever heard of Mixi, or Metroflog, or Yonja?)

The point is, it's still early days yet in social networking, and there are still no guaranteed winners and losers. As our Web Globalization analyst Zia just said to me -- I wonder what this map will look like five years from now?

One of the things I focused on in the report was the integration of specialty search (e.g., image or video search, directory search, shopping search, news search) into standards results pages -- what Google calls Universal Search, and what Ask implemented as part of their 3D search results pages. During my research I found a couple of good blog entries on the topic over at Ramblings About SEO. They did the kind of thing I like to do: they tracked each engines' response to a variety of search queries to see what types of multimedia content they're integrating into their results pages, and how far up the page these multimedia results appear. You can see the results of their image webtrack here, and the results of their video webtrack here.

The bottom line is that it's early days for integrating specialty search into standard search results, and that different engines do well under different conditions. But the results are complex (and interesting), so you should go read through them yourself.