That stuck with me. Relevance, as determined by search algorithms, and usefulness are not the same thing. And then, John Battelle touched on the same topic in a blog post a few months back: "So first, how would I like to decide about my quest to buy a classic car? Well, ideally, I'd have a search application that could automate and process the tedious back and forth required to truly understand what the market looks like."
advertisement
advertisement
Navigating Complex Decisions
Again, this concept of usability comes into play. Let me give you another example. As my regular readers know, I love to travel with my family. But the available travel sites still require the tedious back and forth that Battelle talks about.
We're not big on hotels or restaurants. We love home exchanges or renting apartments and homes directly from the owner. We tend to fly on mileage points. We don't take bus tours, but we do rent cars. We prefer staying in smaller towns rather than big cities. And our first day in a new location always involves a trip to the nearest grocery store.
There is no online destination that brings all the usefulness I need together into one place. I manually pull information from VRBO.com, Homeexchange.com, TripAdvisor.com, Kayak.com and a dozen other sites.
Planning a family holiday is a lot of work, but I'm willing to do it because it's fun for me. What about tasks that aren't as much fun? What about the planning that has no inherent reward, like a complicated purchase for your company, or a forced move to a new city? The title of Battelle's post was "Search Frustration: It's Still Hit or Miss on Complex Decisions." I can relate.
This was the approach Microsoft decided to take with Bing.com, the "Decision Engine." I think their instincts and strategy are right, but the execution is off. If I search for Bristol, England (we're doing a home exchange there next summer) on Bing, I still see a pretty standard search results page. It's not that useful to me.
I agree completely that there's a strong need for more usability in search. Google's Achilles heel at this point is its focus on relevance at the expense of usability. We need a much deeper, more useful experience. Relevance is a poor proxy for usefulness. It still leaves all the heavy lifting up to the user.
Search or Decision Engine? Just Decide!
"Usefulness" is a difficult trick to pull off. It's a tough road that Microsoft has chosen. But if you're going to do it, commit fully to it. Don't play the safe middle ground. This is not the place for half measures.
Whether by design or by luck, I think Microsoft picked the one area where Google is most vulnerable, but right now there isn't enough differentiation between the two. If Microsoft truly wants to be a "decision engine," its strategists have to build from the ground up to offer more usefulness. I'm now four clicks into Bing for "Bristol, England" and still haven't found anything particularly useful to me. Four clicks are way too many. The information forager in me would have already moved on to a new destination.
The next three years in search will be about aggregation of information and incorporating usefulness. Search will do much more than just organize the world's information; it will allow you to do something with it. Search will become the ultimate mash-up. And increasingly, those intersections will happen on mobile devices. Microsoft is the only one of the major players to have declaratively set sail in that direction. My advice? Forget what search is today and move with all possible speed to what search needs to become tomorrow.
I agree! - Good notes on Search or Decision - Just Decide
Achieving the usefulness goal is very abstract. Search engines are cold blooded dead machines. Unless we develop technology that will actually invade our minds and read our thoughts 'usefulness' will never be attained unless we meet 50% of the way. In your example for Bristol you didn't do your part. Plugging in such a simple term that could mean anything the machine will spit out anything from facts about bristol, to the visitors bureau site, to the wiki page. Now if you were to help the cold blooded calculator by adding to your search query much more relevant information comes back. I bet 99% of the time the exact information you ask for does come back in the results, but we expect first page maybe second. But I bet in the first 100 pages of results you will find what you want.
Now a real development would be a Question and Response format. This is perfect for Voice Query Software if we could ever truly master this technology. Bristol, England....then a response asking for more info or guiding you would be perfect. Travel? Information? Business? etc....you say Grocery. It returns with traditional? boutique? organic? quick? You say traditional. And out pops a list.
Search is a means to an end not a solution for demanding lazy people. I am not accusing you of being lazy otherwise you would only travel on pre-packaged bus tour trips! Just trying to incite a discussion of a very abstract idea.
I agree with Howie that you didn't do your part Gord. Just adding one more word to your query gives much closer results to what "you" know what you are searching for. By searching "Bristol, England travel" or "Bristol, England vacation" I am sure you find much more meaningful, and relevant results to what you are seeking.
How can the SERP's become better is at deciphering popular searches related to your generic search. True your search could lead down many paths, and the SERPs are doing better at Universal Search results, but they could do better. With your search as an example, the "related searches" links on the left side of Bing show your usual: map, history, zip code, etc. Maybe not enough people want to travel/vacation in Bristol to warrant a related search link?
As you pointed out, you may not be your a-typical traveller either, which makes the job of the SERPs that much more difficult in figuring our true intent with broad searches. Enjoy your travels next summer, and hoist one "to Arthur" for me!
I enjoyed reading your post. While doing so, the concept of negative keywords or filters came to mind. Anyone who uses Google AdWords understands how they increase the relevance and usefulness (thus efficiency).
Another everyday example of how negative keyword filters can make a query more accurate is an advanced EBay search.
In both examples above, we are able to choose the negative keyword filters we want and save the settings. It seems like this concept could be employed to get better search results in other situations, such as the one you described.
Your search also would be a lot easier without the repetition, repetition, repetition, repetition ... AWK ! .....of the same sites and your first 100 pages could come down to 20....from just a user.
Gord, loved your insight from the world of travel. The experience you describe is precisely the reason we started NileGuide---to solve the frustration of finding relevant content through traditional means, and then the added challenge of actually doing something useful with that content (like planning a trip, for example). Check us out and let us know what you think.
Josh Steinitz
CEO
NileGuide
I agree that it is very hard to find the right content on the internet. I was looking for a key west bed and breakfast and I got so many terrible sites that came up. I wish there was something that could be done about it.
I am going on a vacation to orange county, california next weekend and I am looking for orange county hotels. I've been trying to find some online but I'm having the same problems as everyone here. Any suggestions?