Monday, November 06, 2006
Search as Shopping
I was recently struck by a thought as I was browsing through the aisles of my local supermarket: how much search was like shopping. Let me clarify: when you go into a supermarket, you either know what you're getting, just want to browse, or maybe a little of both. Sound a little like search to you?
Search can be motivated in three ways (PDF): by information, transaction, or navigation. Information meaning "I want to find out more about polar bears". Transaction meaning "I want to buy the latest Harry Potter book from Amazon at a low price". Lastly, navigation meaning "What's the address for Google?" (don't laugh, this is actually more common than you think). Like shopping, you're motivated by both visceral and whimsical factors in some proportion every time you do a search.
So, what's the best way for you to actually find what you're looking for? Traditionally, a search engine just gives you a list of results with a few key phrases highlighted, and you have to do the hard work of figuring out if a site is what you're looking for. Contrast that with shopping where you head toward the product you're looking for via very generic markers (a placard above every aisle listing what's there).
Let's step back a bit, and start at the beginning of our journey into the supermarket. Notice that your typical supermarket has a bunch of items stacked near the entrance. You can't get in the store without at least glancing a little at what's displayed. If you were ever going to impulse buy something, it's in those first few seconds when you step in the door. There are no other goods competing for your attention. It's like tunnel vision, channeling your line of sight. Like search, the first page of search results is what you look at most closely. Maybe it's the only page you look at. Even if you're the rare, determined searcher who isn't content until you find what you're looking for, the sheer drudgery of skimming through page after page of results wears you down.
Back to the aisle. Once you actually get to the aisle, you don't usually look at every single box of, say, coffee to find the one you're looking for. If it's your favorite brand, all you do is look for the telltale brand and coloring of the package. In other cases, a new item, because of its novelty, catches your eye. You don't recognize it so you look closer. Notice that you don't really pay attention to the boxes you know aren't what you want. It's all handled subconsciously.
Of course, it's the fact that Folgers has done a lot of promotions that you recognize its brand. Not so when you're searching the web where it's unlikely you'll recognize any page you're looking at. But, think, what if a search engine automatically annotated search results with little logos? Say, the results that lead toward a page hosted by Yahoo gets a little Y! logo next to it. Let's go further, what if, for those pages located in the long tail of the web (the ones you would never have known existed if you hadn't used a search engine to look for it), a small screenshot of the site or some other visual cue were placed next the result?
You've probably noticed that it doesn't take as much effort to take in a picture or painting as compared to reading a novel. A picture is worth a thousand words, as the saying goes. The cognitive load is lower to look at a picture than to read a piece of text. It's the classic left brain/right brain divide:
Search engines like Google and Yahoo appeal to the left brain with lists of search results. Doing a little research, I found several less popular search engines that appeal to the right side of the brain like Snap.com (with thumbnails of sites next to search results), newcomer Quintura (with a tag cloud of suggested keywords), and Grokker (with its set theory-inspired clusters).
Are these search engines actually better? Personally, I don't find right brain search engines to be better than Google, but others with different sensibilities might disagree. If you need to quickly analyze a set of data, the proper visual representation of a large dataset is critical. In that case, a right brain type of search engine would be very helpful. As far as the current crop of right brain search engines, I'm still shopping for better. Much more work needs to be done to truly leverage the power of pictures.
Search can be motivated in three ways (PDF): by information, transaction, or navigation. Information meaning "I want to find out more about polar bears". Transaction meaning "I want to buy the latest Harry Potter book from Amazon at a low price". Lastly, navigation meaning "What's the address for Google?" (don't laugh, this is actually more common than you think). Like shopping, you're motivated by both visceral and whimsical factors in some proportion every time you do a search.
So, what's the best way for you to actually find what you're looking for? Traditionally, a search engine just gives you a list of results with a few key phrases highlighted, and you have to do the hard work of figuring out if a site is what you're looking for. Contrast that with shopping where you head toward the product you're looking for via very generic markers (a placard above every aisle listing what's there).
Let's step back a bit, and start at the beginning of our journey into the supermarket. Notice that your typical supermarket has a bunch of items stacked near the entrance. You can't get in the store without at least glancing a little at what's displayed. If you were ever going to impulse buy something, it's in those first few seconds when you step in the door. There are no other goods competing for your attention. It's like tunnel vision, channeling your line of sight. Like search, the first page of search results is what you look at most closely. Maybe it's the only page you look at. Even if you're the rare, determined searcher who isn't content until you find what you're looking for, the sheer drudgery of skimming through page after page of results wears you down.
Back to the aisle. Once you actually get to the aisle, you don't usually look at every single box of, say, coffee to find the one you're looking for. If it's your favorite brand, all you do is look for the telltale brand and coloring of the package. In other cases, a new item, because of its novelty, catches your eye. You don't recognize it so you look closer. Notice that you don't really pay attention to the boxes you know aren't what you want. It's all handled subconsciously.
Of course, it's the fact that Folgers has done a lot of promotions that you recognize its brand. Not so when you're searching the web where it's unlikely you'll recognize any page you're looking at. But, think, what if a search engine automatically annotated search results with little logos? Say, the results that lead toward a page hosted by Yahoo gets a little Y! logo next to it. Let's go further, what if, for those pages located in the long tail of the web (the ones you would never have known existed if you hadn't used a search engine to look for it), a small screenshot of the site or some other visual cue were placed next the result?
You've probably noticed that it doesn't take as much effort to take in a picture or painting as compared to reading a novel. A picture is worth a thousand words, as the saying goes. The cognitive load is lower to look at a picture than to read a piece of text. It's the classic left brain/right brain divide:
Following the standard line, the left hemisphere is the logical hemisphere, involved in speech, reading, and writing. It is the analytical hemisphere that evaluates factual material in a rational way and that understands the literal interpretation of words. It is a serial processor that tracks time and sequences and that recognizes words, letters, and numbers. The right hemisphere is the intuitive, creative hemisphere. It gathers information more from images than from words. It is a parallel processor well suited for pattern recognition and spatial reasoning. It is the hemisphere that recognizes faces, places, and objects.
Are these search engines actually better? Personally, I don't find right brain search engines to be better than Google, but others with different sensibilities might disagree. If you need to quickly analyze a set of data, the proper visual representation of a large dataset is critical. In that case, a right brain type of search engine would be very helpful. As far as the current crop of right brain search engines, I'm still shopping for better. Much more work needs to be done to truly leverage the power of pictures.
Technorati Tags:
Labels: Technology
