Search: Social Or Algorithmic?


Many of us spend large parts of our day searching for information. After attending Social Web FooCamp over the weekend, I’ve been thinking about how I search for information using a combination of social and algorithmic methods. When I talk about “social search,” I’m referring to using your network or other human beings to find information. In essence, you’re relying on the knowledge of other people, instead of using Google (s goog) or another search engine, which uses algorithmic methods to determine the best results. In Sam Dean’s recent post, “5 Free Online Answer Sites for Tech Questions,” he shared five sites where you can ask questions and get answers from people with subject matter expertise, which is a social approach to finding information.

Algorithmic searches, on the other hand, are great for finding information when you know enough about a topic to formulate a strong query that will return highly relevant results. These searches work best for me when my question has little ambiguity and when I know enough about a topic to formulate a good search query.

Aardvark Social SearchOne big advantage of using social search is that you can find information when you aren’t quite sure what you need. Social search approaches allow you to describe a problem or ask a question that isn’t formulated well enough to put into a search query. This is useful when you don’t know enough about a topic to know where to start, or when the algorithmic results are returning questionable results. In an earlier post about research tools and techniques, I discussed an example of using my Twitter network to get an answer to a question about event registration systems. While this is great when you have a network of people who might know the answer to your question, it won’t work well when you need to know about a topic that falls outside of the expertise of your personal contacts. However, new social search services like Aardvark are working to fill this gap to include answers from friends of friends.

I’ve found that I get the best information if I start with my social network to get a few suggestions and then take those suggestions to Google or other search engines to get more details and additional information. By getting a good baseline of information from friends and other people in my network, I can refine my algorithmic search approach to get more relevant results. This combination of social and algorithmic search has worked well for me.

How do you search? Do you use both social and algorithmic search methods?


Dawn Foster


I agree with Brynn that it isn’t an either / or question. I use both, but for very different things, and I often use them together. I go to Google first when I know what I’m looking for, but I turn to social search when I need the human touch (gadget suggestions, getting started on a complex topic, etc.)


Its nice to get a brief outline about a certain topic and then search for further details with appropriate keywords in major search engines to get refined results.
Great tool I beleive!!

Brynn Evans

@Anelly Social search and algorithmic search aren’t mutually exclusive! They’re both useful to searchers but potentially for a different types of questions. That’s what makes Aardvark so powerful—questions you ask there can’t easily be answered with Google (or require a lot of foraging on Google).

I find it interesting, Dawn’s comment, about going to social networks to get suggestions first and then taking them to Google. This is one thing people are really good at: framing a question and providing ideas on how to scope it. In fact, there are social/collective intelligence ways of algorithmically helping searchers here in similar ways—it all says to me that social & algorithmic methods need to be better integrated online.

Thanks for your post!


it seems that Aardvark it’s a great tool for getting the right answer. It’s great but i’m not prepared to give up to algorithmic research yet.

Rob Spiro

Thanks for the great writeup! Very cool to hear that Aardvark is helping you out as a complement to Web Search — we’re finding that it’s extremely useful in those scenarios.

Comments are closed.