-->

A New Kind of Search Engine

March 26, 2010 

Andrew McAfeeBetween South by Southwest and the Cognizant Community 2010 Conference, I’ve heard some fascinating presentations over the last couple of weeks. I want to tell you about one in particular, though, because it introduced me to whole new ideas about how we acquire information.

The speaker was Andrew McAfee, principal research scientist at the MIT Sloan School, fellow at the Harvard Berkman Center and author of Enterprise 2.0. McAfee specializes in the application of advanced Internet technologies to corporate communications, and his observations about the impact of Twitter and Facebook on the way we find information raise the possibility that a new kind of search is emerging.

Speaking at the Cognizant conference earlier this week in Scottsdale, McAfee described how much the process of finding information has changed in just the last 15 years. As recently as 1995, the most common reference source we had was a library where professional human curators made decisions about what we needed to know. Information was not only scarce but constrained by space and the limitations of indexing systems that forced information into uncomfortable categories (David Weinberger’s Everything is Miscellaneous describes this brilliantly).

When the Internet went mainstream, we initially tried to recreate the curated model online. Remember that Yahoo started as a structured taxonomy designed by humans that organized the Web into categories. There is some value to that, but few people access information that way today.

Instead, we discovered that search engines are faster and bring us directly to the information we’re seeking. It’s amazing how quickly people have discarded the library metaphor that dominated our thinking just a decade ago in favor of search. In December, people conducted more than 4.7 billion searches worldwide every day.

A New Approach to Search
Now there may be a new kind of search taking shape based upon the ask-and-answer principles introduced by social networking. Twitter users understand this well. Let’s say I’m in Chicago looking for a place to take business colleagues to dinner. I can search the Web for restaurant reviews, but I can also ask a question of my followers: “Recommend a good restaurant within 10 minutes of McCormick Place?” Both actions yield useful information, but the Twitter inquiry may actually provide superior value because the response comes in real time from people I know and trust.

I’ve already noticed my behavior changing as a result of this network effect, and perhaps you have, too. When I’m about to make a major purchase decision, I often ask my Twitter followers for advice. In effect, I’m conducting a search against a database of unpublished information that’s stored in people’s memories.

If we can unlock and share this untapped resource, we can potentially open a treasure trove of new information. In McAfee’s words, “Your ignorance makes everyone smarter.”

Organizations that are experimenting with Web 2.0 tools behind the firewall are discovering that this is a remarkably powerful idea. For 20 years, we’ve tried to capture knowledge by interviewing veteran employees and storing what they told us in databases. That’s never worked very well because it’s an unnatural knowledge-transfer mechanism.

It turns out that people are more generous and spontaneous with expertise when they answer ad hoc questions from peers. Some organizations are beginning to scrap the old tools in favor of this free-form exchange.

The trick is to preserve, organize and rank this wisdom. You can bet that Google and others are trying to figure that out right now. I was a little mystified last month when Google acquired Aardvark, a “social search engine,” for a pricey sum of $50 million. Aardvark is sort of a structured Twitter; its members can ask questions of others who have a self-declared area of expertise.

Having listened to Andrew McAfee’s insights, I now understand better what Google executives were thinking. This doesn’t mean that today’s search engines will become irrelevant. Social search is an extension of an already-powerful metaphor, and it has some very exciting implications.

What do you think? Are there scenarios in which social search could replace the ubiquitous Google query box?

Free Webinar This Week, Then Meet Me In Chicago

Just a reminder that next Thursday I’ll conduct a free seminar from Awareness discussing the interim findings of some research I’ve been doing into business use of multiple social media platforms. This research indicates that there’s been a striking increase in just the last couple of years in the number of platforms social media platforms that businesses are using. Click on the link to sign up.


On April 8, I’ll be in Chicago for a luncheon that the local Business Marketing Association is sponsoring on how to do business marketing on Facebook, Twitter, YouTube and LinkedIn. Each speaker will cover a different platform, and I was lucky enough to draw the Twitter card. Hope to see you there!

Tip of the Week: Dlvr.it

Dlvr.it entered public beta test this week, and it’s a tool worth looking at if you’re interested in getting more mileage out of your publishing activities. Dlvr.it distributes information from your RSS feeds to Twitter and Facebook without user intervention. Other services will be added soon. So you don’t have to tweet the stuff you publish; dlvr.it does it for you.

The real power of the service, though, is on the back end. For each syndicated item, whether it be an article or a single tweet, you can see who retweeted the item, how many times it was distributed and how many clicks it received. There’s a metric called “direct reach” that measures the follower count of people who re-post an item. There’s also a calculated metric called “extended reach” that figures out how many people have tweeted your content using other URL shortening services. You would have no other way of knowing about this activity other than through by looking at server logs for referring URLs.
The service still needs some work on the user interface side, but the back-end metrics appear solid. I’ve seen a noticeable uptick in retweets of my blog entries since I started using it.

Just For Fun: OK Go’s This Too Shall Pass

You might be familiar with OK Go, a band that is known for its inventive and intricate music videos. They’ve outdone themselves with their latest effort, a four-minute celebration called This Too Shall Pass that features some of the most innovative ideas I’ve ever seen in a Rube Goldberg machine. It ends with paint guns being fired at the band members. Considering the amount of paint the the band members are wearing at the beginning of the video, it no doubt involved many takes. But the results — nearly 10,000,000 downloads so far — are probably worth it.

OK Go's This Too Shall Pass

Comments