BrandSavant

Gaining Insight From Social Media Data

Back To The Browse

by Tom Webster on June 4, 2010

Some years back, I attended a talk by Dean Hachamovich, who heads up the IE team at Microsoft. Dean recounted what he saw at the time as the three phases of the Internet. We began with “Browse,” when we all got our first AOL accounts and search wasn’t very good. We killed a lot of time this way.

Then, search got a whole lot better, and we didn’t have to rely on vertically-scrolling monstrosities of portal pages to find what we needed. This new era, the “Search” phase, made things much more efficient, as directed information-gathering replaced much of the inefficient browsing we used to do for information.

I think I saw this talk back in 2005, which is when Dean indicated we were on the cusp of the next, third phase of the Internet – “Subscribe.” Though he saw search continuing to be the dominant paradigm in the near term, he believed that continuing development of RSS and other tools would empower consumers to search once, and subscribe to those searches to get information, which would be an even more efficient way to interact with the Internet than Search.

A funny thing happened to that model in the past couple of years, however: social media. About half of all Americans have a profile on one or more social media sites, and while subscribe remains one of the dominant paradigms, it has morphed from subscribing to information streams into subscribing to people.

I thought about these models today when I saw a tweet from Jeff Jarvis about Bit.ly’s recent growth, and how more Bit.ly links were decoded in May (4.7 Billion) than Google links to publishers (about 4 Billion). It struck me that when I click on a Google link, it’s because I searched for something. I knew exactly what I wanted, and I’m good at search. A lot of us are, now.

What I do when I click on most Bit.ly links, however, is something entirely different. If you think about the URL-shortened links you typically click on, I bet you’ll find that you come across these the same way I do – on social media platforms, through some kind of sharing mechanism. In short (pun intended), you found them while browsing.

In truth, what advances in search and subscribe have really done for us is to give us better filters for browsing. We like to browse. In fact, we need to browse. True, the ability to search for and find exactly the information we need is crucial. But what about the information we don’t need – or better, don’t yet know that we need? My wife the scientist tells me that the only difference between basic science and applied science is that basic science just hasn’t been applied – yet. Making serendipitous connections, seemingly at random, is how many breakthroughs really happen. What precedes great discoveries often isn’t “Eureka!” but rather, “that’s funny…”

In my world, we call this “Undirected Knowledge Discovery.” Data Mining. You might call it “browsing.”

Long live the browse.

Buffer
  • Pingback: Tweets that mention Back To The Browse | BrandSavant -- Topsy.com

  • http://www.grassshackroad.com mike mcallen

    As I was rendering some video I decided to do some browsing and catch up with my snarky comments to my favorite bloggers. I have nothing for you today Tom. I think you are spot on with this. As I have tried to stay ahead of the curve (unsuccessfully much like my baseball career) but I was talking with a friend who is a PR guy just entering into the social media realm. He also pointed out Bit.ly to me and the info it provided. I am a terrible master of my schedule and bit.ly links along with twitter has made me a browsing fool. Not just a fool. Thanks for the post.\nm

  • http://www.outcome3.com Jon Paul Janze

    Interesting Article, I think rather than ‘browse vs search’, there are situations where either or both behaviors are appropriate. Peter Morville, discusses the evolution of search behaviors and solution designs extensively and demonstrates that browse is often more appropriate for low complexity, low option environments and search is often more appropriate for areas of ultra high option environments, but a blend of the two, what is known as faceted search, wherein search boxes are connected to criteria filters etc, is probably most appropriate in these situations.

Previous post:

Next post: