Name a small, noisy bird. A chipmunk. At what month of pregnancy does a woman start to look pregnant? September. What is a man’s name that starts with the letter K? Kentucky Fried Chicken. Something that a burglar would not want to see when breaking into a house? A naked grandma.
These are real answers, folks, given by subjects in a study of free association published by the ABC Television Network under the title Family Feud. As you know, to play the Feud, a genial host asks contestants a question with an ever-so-slightly lewd undertone, at which point the pressured contestant tries to come up with the G-rated answer most given by one hundred non-pressured people who were polled ahead of time. It’s the same skill that at work can make the difference between a five-second Google search and ninety minutes spent browsing for the information you need. Think about it. How many times in an average workday do you search for information online? And how did you learn to perform these searches? If you’re like most of us, the answer to that last one is trial and error. Now’s the time to bring just a little bit of science to your Internet search strategies.
First, there are a ton of search strategies to choose from, many of which you’re probably already using without knowing you’re using them. Take a minute to see which of the following strategies, described in a 2012 paper in the journal Information Processing and Management, you already use in your searches:
• Identifying search leads to get started (e.g., deciding to search CNN.com rather than all of Google)
• Creating a search statement
• Modifying the search statement (e.g., if “Why does my operating system hate me?” was too broad, you might try “Why does Mac OS 10.7.5 hate me?”)
• Evaluating search results
• Evaluating individual items
• Keeping a record
• Going forward
• Going back
• Learning (e.g., learning how to use a more advanced search interface)
• Exploring
• Organizing (e.g., sorting results by date instead of name)
• Monitoring
• And, finally, using
Here’s the important part: some of these strategies work better than others. We know what works because researchers have compared the behaviors of expert searchers to those of novices. Here’s the scoop: expert Internet searchers spend much more time than novices evaluating search results, going back, and organizing search results—it’s as if they do more with fewer search terms, getting intelligently into a “search ballpark” and then working systematically from there (as opposed to throwing many hopeful searches against the wall and seeing what sticks). Here’s another cool little expert nuance: “evaluating search results” isn’t clicking individual links to see if they give the information you want, but looking at the quality of the list of results as a whole. Did the search terms get you in the ballpark? If so, might you be able to tweak the search terms to narrow that ballpark? Finally, experts know not to chase white rabbits down holes—if a click takes them further away from rather than closer to their information goals, experts go right back to the most recent page that was on the right track. Don’t be afraid to click that back button, and then stay organized in a way that keeps you from going down that dead end again.
Taken as a whole, the process of expert Internet search is much like the process of problem solving itself: an efficient Internet search requires time spent on the front end designing the problem-solving strategy, and much less time spent actually executing this strategy. Use your powers of Family Feud free association to search the terms that most people associate with your information. Tweak your terms to refine the results. And when in doubt, go back rather than forward. How many minutes or hours do you think you can save in the course of a forty-hour workweek by dialing in your powers of Internet search?
Do you want to test your ability to win Family Feud … and in the process get a little better at picking Internet search terms? Check out the University of South Florida Free Association Norms database (luckily, searching exactly those words will get you right there). Basically, the database gives you a stimulus word, like dinner, and then shows what words people tend to associate with it, like supper and eat. It also shows how strongly we associate these words with the stimulus. For example, if we go back to Family Feud, if the stimulus word is dinner, then supper would be a bit higher on the list than eat. Try it yourself. Head to the database and play around. How closely do your associations match the associations of people polled for this sophisticated version of Family Feud?