Late last year, in my predictions post for 2007, I mentioned something I called, quite uninventively, “Blog 2.0.” More specifically I wrote:
“10. “Blog 2.0″ will become a reality. By this I mean that Version 1.0 blogsites, of which I think Searchblog is a good example, will begin to look dated and fade in comparison to sites that employ better approaches to content management, navigation, intelligent widgets and web services, etc.”
Well, that entry caught they eye of Martin Nisenholtz, head of NYT Digital. I was in NYC this past couple of days visiting colleagues and attending the wonderful “Evening of Wonders,” run by FM partner author Josh Foer. Upon hearing I was coming to town, Martin asked me to stop by. He wanted me to unpack what I meant by that brief reference. Turns out, we had a fascinating conversation, and it led me to want to think out loud with all of you about a problem I think most folks who either write or read blogs have.
Blogs are wonderful for reading a stream of consciousness, the ongoing dialog of the author(s) and the audience (though I have to say reading comments is far more of a pain that it should be.) One of the major strengths of blogs is their conversational immediacy, when one becomes a reader, and knows the flow and grammar of the site, the ongoing postings become far more accessible.
A brief dip into nearly every blogger’s referral logs shows that a very large percentage of readers – nearly 40 percent in some cases – come directly from search – someone who put “steve ballmer throws chair” into Google, for example, and lands here.
Now, this person doesn’t have any frame of reference about Searchblog, or its grammar, audience, or ongoing conversation. He or she is most likely to hit the post in question, read it (perhaps), and move on. This site loses a potential new reader, and this community loses a potential new member, because, in the end, I, as the publisher of Searchblog, have done nothing to demonstrate to that reader the wonders and joy that is Searchblog. In short, I’ve failed to merchandise my site.
Now imagine instead, that when that person comes from search, they are greeted with a box that pops up and is informed by the search referral information that we all carry with us as we click away from Google or other search engines. That box surfaces a smart search based on the referral – perhaps it shows the reader other posts I’ve written about Microsoft, or Google and Microsoft, or senior executives in the Internet industry. Perhaps it shows me the top five *other* posts folks read who *also read* that Steve Ballmer Throws A Chair post. You know, the kind of merchandising a good site like Amazon does all day long (from what I can tell, search referral boxes were pioneered by Cnet, for credit where credit is due). Now, wouldn’t that be cool, just for a start? Sounds hard to do, right? But actually, it shouldn’t be. The information is all there. It’s just not organized properly.
But giving potential new readers a tour of Searchblog is not the only reason to think about merchandising the site. In fact, it’s not even the primary one. The real reason to think about solving this problem is to solve what I as a regular reader of many sites find most frustrating: it’s nearly impossible to navigate these sites past the home page.
Sure, some sites, including this one, make lists of “most active topics” or “my favorite posts” that we stick on our templates. These static navigation boxes help a bit, but they fail utterly to solve for a reader’s implied or specifically declared intent. And finding anything specific on a blog site is, well, a pretty poor experience. I’m not saying blog search technology sucks, I’m saying it simply doesn’t work very well. Often you find far too many posts for the keywords you enter, and there’s really no context for the ones you do find.
Now, I know what many of you are thinking – this is the problem that tagging will solve. OK, sure, let me know when that happens. For now, tagging strikes me as one more signal lost in a ton of noise – it’s up to publishers to take that signal and make something useful from it.
And as long as we are talking about signals, there are so many other ones I’d love to have at my fingertips, signals I as a publisher can weave into useful context and navigation for my readers. Links to a particular post, for example. Why can’t I have, easily, a list of all my posts sorted by how many links each has, or by a matrix of links crossed with authority of those links? How about the number of comments? Or the number of pageviews each post has received? With all of these signals, plus tags, I can start to really build useful navigation elements for all my readers, past, present, and future.
But it’s way too hard to get all this information and harder still to know how to use it. I know it’s out there, in bits and pieces, but it strikes me that no one company is really motivated to address this problem in a way that benefits bloggers. I have high hopes for Feedburner, now that it bought BlogBeat, and perhaps I’m missing some dead obvious widgets that do all this and more for any particular site. For that reason, I’d love to work on this issue with what resources FM has to throw at it. But first, I need to get smarter on the possible solutions. What say you, Force of Many?