The Noisy Channel

 

Why LinkedIn Frustrates Me

March 31st, 2009 · 3 Comments · General

Let me start by saying that I really like LinkedIn. I use it as everything from a self-updating address book to a gateway to professional communities like the enterprise search professionals group. I am delighted by the information LinkedIn has assembled about companies just by aggregating user profiles. In short, I take LinkedIn quite seriously as a professional networking tool.

With that preamble out of the way, I’d like to vent a bit about LinkedIn’s approach to search. Directories are a poster-child domain for faceted search. LinkedIn specifically has high-quality semi-structured data, since users are personally incented to optimize their own findability. Moreover, the process doesn’t even seem adversarial–I haven’t seen any Joe the Scammer claiming to be the CEO of a Fortune 500 company (oops, bad example). LinkedIn has done the best job I’ve seen of aggregating high-quality data about people’s professional history–in a volume that is not only unprecedented but more importantly is large enough to be broadly useful. And the site designers clearly care about search: they still proclaim above the search box, “New Improved Search!” (see my earlier review here).

Why is faceted search so valuable for directories? Because finding someone is often a task that calls for exploratory search. Unless you’re looking someone up by name (and hopefully by a unique name), you’re not performing known-item search. Rather, you’re looking for a potential employee, employer, business partner, or expert. You may not even know what you want until you have a sense of what’s out there–the different companies in the space, the different relevant job titles, etc. Moreover, now that LinkedIn has a significant amount of text associated with itsĀ  users (e.g., the Q&A section and forums), it could do a much better job of linking people to the content they produce.

I understand that uniting the social network functionality of LinkedIn with search is hard enough, and that introducing faceted search makes the problem that much harder. But it’s not impossible, and the value of such an application more than justifies the effort. So far, LinkedIn has benefited from having the best data. But users have no incentive to give LinkedIn exclusive access to that data. LinkedIn knows this, and its increasing emphasis on community will surely make it harder for someone to compete just by offering better information seeking support.

Still, the core value proposition of LinkedIn is tightly bound to the site’s search functionality, and LinkedIn would do well to take a more modern approach. Doing so would increase the site’s value dramatically, and I’m certain LinkedIn would find ways to monetize that value.

3 responses so far ↓

  • 1 Andrew // Apr 2, 2009 at 4:26 pm

    This post is right on the money; as a recently unwillingly unemployed person, LinkedIn becomes a more critical resource than ever before. It is very frustrating to stumble through people’s networks without any capability to navigate or discover options that present themselves based on the data. For example, since I’m interested in breaking into the marketing space, it would be extraordinarily helpful to categorize contacts by (sub)verticals within a person’s network. Or even consider contacts who have X number of years of experience who could be good mentors to be introduced to as part of an industry search. There are a lot of good user scenarios out there waiting to be built out.

  • 2 Daniel Tunkelang // Apr 2, 2009 at 5:55 pm

    Well, I know that people at LinkedIn have heard my ranting, and I can only hope that they are working to address some of these concerns. As I said in the post, directories are a poster-child domain for faceted search.

  • 3 LinkedIn Rolling Out Faceted Search! | The Noisy Channel // Jul 15, 2009 at 10:19 pm

    […] that I known about this project for a while–they’d reached out to me after I offered a lukewarm review of their search–but I was asked to be discreet about that […]

Clicky Web Analytics