Appendix B. Keeping Up
The following pages were originally published online as bonus material for our readers and are not included in the print version of this title.
What I’ve tried to do with Information Trapping: Real-Time Research on the Web is to give you the framework for an inexpensive way to identify and track topics in which you’re interested, and put the results in a form that’s presentable to others. Following that framework to a “T” is a tall order. You may decide you don’t want to set up every possible monitoring permutation in this book. That’s fine. It’s better to set up a partial program that you can keep up with, rather than a full program that quickly overwhelms you.
This bonus chapter looks at ways you can hone the practice of consistently keeping up with your information traps—how to instill sound monitoring habits, decide when to change the keywords you’re monitoring, keep abreast of new sources to monitor, and reevaluate your sources.
Unfortunately, a lot of this guidance boils down to a “gut feeling.” I can’t tell you, “When you’re only getting x alerts a month for this keyword, switch to keyword y for additional relevance.” I can point out symptoms of things that might be going wrong, reasons you might be feeling uneasy, and little clues to which you might want to pay attention. But in the end, you are the best judge of whether your traps are working.
As you get more and more experienced with information trapping, you’ll learn you can very much rely on your instincts to determine when you’ve got it right. This kind of thinking reminds me of the book Blink, in which author Malcolm Gladwell talks about how people can make decisions in just a fraction of a second, based on a lot of subconscious cognition that they might not be aware of. As you monitor your topic, as you read about it, as you write about it, and as you organize information about it, I suspect this will happen to you as well.
But it doesn’t happen magically. First you have to get in the habit of monitoring every day.
Brush, Floss, and Trap
How much time you spend every day monitoring the Internet for information really depends on how many topics you’re monitoring and how much news they generate. But even if you’re only watching for news on antique woodworking tools, it’s best to do it every day, so that you get into a rhythm of checking and you get a sense of what’s normal in terms of how much information your traps routinely uncover each day.
I used to spend mornings checking my traps; now I tend to check them in the evening. It seems to work better when I do it all at once instead of over the course of a day—it’s easier to catch duplicates, get a sense of trends in coverage, and so on. I go through the RSS feeds and monitored pages separately and manage the e-mail alerts as I get my e-mail.
For every item I get, I try to do one of three things:
• Put it in a text file to be acted on.
• Pass it on to someone who will be more interested in it than I am.
• Throw it away or ignore it.
The temptation is always there to save something for later, of course, but I try to resist it. Virtual objects have a tendency to pile up and become useless because you never have the time to actually read or act on them (at least they’re findable via computer indexing programs). If I absolutely must keep something, I try to relegate it to a storage text file that I don’t normally use, but which I can easily search if I suddenly have a need for some stashed story.
If you decide to put off dealing with some of your gathered information in this way, be sure to save entire articles rather than just pointers to articles. Some sites don’t keep their news archives up for more than 30 days or so, and you might find yourself looking for a story that was relegated to a pay-per-view archive months ago. Full-text articles also make searching a little easier as well.
So you want to be monitoring every day, at a regular time that suits your schedule. If this isn’t possible, commit to checking your traps at least three times a week. By developing a regular habit of checking your traps, you’ll get a sense of the cycle of news for your topic. In the case of my trapping topics, weekends and holidays are really slow; Monday and Tuesday are torrentially busy, and the rest of the days are busy enough. Things may be different for you. If you’re tracking a particular sport, for instance, you may have very busy weekends and then months when very little is going on because of the off-season.
Knowing the average day-to-day activity of your topics is what allows the gut feeling to come into play. For instance, sometimes I’ll notice that a weekend has been a little busier than usual—a rumor here, a story there—and then I start wondering if something’s going to come down early in the week. Often I’m right. Other times I’ll notice a certain story repeating over and over and over again, and I’ll wonder if that means the story is suddenly going to change in a dramatic way, and often it does. Once you’re used to the rhythms of your topic, you’ll have a “Hey, what’s up?” feeling when you’re suddenly getting a lot more, or a lot fewer, alerts.
When you get that instinctive feeling about a topic you’re trapping, you definitely want to pay attention to it. It’s that feeling that sometimes lets you know it’s time to make changes to the keywords you’re monitoring.
Know When to Hold ‘Em, Know When to Fold ‘Em
Most keywords for which you’re monitoring are going to be nouns. (I can think of a few times when you’d use verbs and other parts of speech, but let’s stick to nouns.) And a noun, you may remember from your Schoolhouse Rock, is a person, place, or thing. How often you will need to change your keywords and what to look for really depends on whether you’re monitoring a person, place, or thing.
Trapping a person
A lot of people monitors are centered on companies—you want to see what the CEO is up to, what the CFO said last week, and so on. It’s easy to know when to alter these—when a person leaves the company, is moved out of the position you wanted to monitor, and so forth. Sometimes I will leave a person monitor in place if the individual left the company in an extraordinary way. If he or she was prosecuted for criminal activity and forced to resign from a company, I may continue to monitor the name a little longer. I do this because that person’s actions most likely will “echo” in the company for a while, and will be referred to again and again as the company deals with things like investor lawsuits, restructuring, and realigning its business.
You might also monitor a name in the context of a topic. Take Peter Drucker, for example. Because Drucker is a famous name in the world of business, you might have added his name to a monitored source about corporations to glean quotes from him, references to his books (many of which are considered classics in management), and so on.
After Drucker passed away in late 2005, you would have to reevaluate your choices at that point. You might decide to remove his name from your monitoring, because you were looking for quotes by people considered expert in the world of management. On the other hand, you might decide you want to keep his name in your monitoring, because he’s often referred to and you like the quality of the results you get when you include his name.
You may also occasionally run across a person who’s written a book or has otherwise become famous in a field you’re monitoring. In this case, you’ll probably see the name in your monitored results for a while, and then at some point you will realize that all the best results have this person’s name in them. That’s your cue to either revise your results to include this person’s name or add a new monitor with just the name and possibly one general keyword for the topic.
In summary, there are a few specific events that might trigger you to remove a name from your monitoring (such as a death or removal from a company), but adding names to your monitoring will be more a case of increasingly finding those people mentioned in your monitored results and then deciding that adding their names to your searches will generate only a positive impact on your search results.
Trapping a place
Places aren’t used very often as keywords. You search within a geographical area for keywords, but you generally wouldn’t find yourself searching for the name of the geographical area itself.
In my experience, places become notorious in a good or a bad way, and this reflects in monitoring results. Silicon Valley is synonymous with the information technology industry, and can be used as a good query phrase when you’re trying to monitor IT in California. On the other hand, if you’re trying to monitor for information about weather disasters, "New Orleans" is going to trigger a huge number of results, commentary, and analysis.
Positive associations tend to develop slowly, while negative associations can appear overnight and will have a huge impact on your monitoring results. And you’ll see them—you won’t be able to miss them! You will have to do some experimenting with location names in your search results to see how much difference they make to your monitoring, and whether it’s helpful or not.
Trapping a thing
“Things” probably make up the bulk of your monitoring keywords, and that’s as it should be. In many cases the “thing” keywords can change often, too, depending on whether you’re using proper nouns or not.
Proper nouns
Proper nouns, as you may remember from English class, are nouns that name a specific person, place, or thing (we’re going to focus more on them as things).
There are lots of proper-noun things that act as great query words: brand names, company names, specific product model numbers, and so on. If you can add a product name or model number to your search, it does a tremendous amount for narrowing your results. The problem is that model names and numbers tend to change a lot. You have to be very vigilant about making sure you use the latest model names or numbers in your query.
Take, for example, the iPod Mini. In 2005, it was superseded by the iPod Nano. If you were trying to monitor for information on flash-memory-based music players and didn’t change your query from "iPod Mini" to "iPod Nano" you’d be missing out on a lot of potentially good stuff. You always have an opportunity to change and update your queries as you trap press releases, reviews, and product updates, so you need to stay vigilant about these types of changes.
Brand names and company names don’t change as often. Sometimes they do (KFC was once Kentucky Fried Chicken), but when that happens there’s a lot of hullabaloo about it and you’ll get plenty of opportunity to change your queries. The model names/numbers and brand names are what you have to watch for.
Common nouns
Common nouns are nouns that are not specific, such as tree, cat, bottle. You can often use common nouns in conjunction with other parts of speech to create narrowly defined query phrases, such as "Antique woodworking tools".
In my experience, common nouns do not change a whole lot once you’ve done your initial experimentation and decided on your queries. You may change how you describe a noun ("East European woodworking tools"), but often you won’t need to change the noun itself. The big exception is technology topics. Tech people love to invent words or use words that have other connotations. Nobody knew what a podcast was three years ago—now it’s an incredibly common word. The word “mashup” was used to mean a combination of songs. Now it also means a combination of data from Web sites combined into one site or tool.
A typical cycle for a tech word can start like this: you see it mentioned on a few blogs. Then you see it mentioned on a ton of blogs. Then you see tech news outlets explaining what it means in their stories. Then they stop explaining what it means because everybody knows what it means. Then your grandma starts using it all the time. Okay, maybe not that last one.
If you see a new word that you think would be useful to your query, you should aim to add it when it starts appearing regularly in a ton of blogs. Start including it too soon in your searches and you’ll narrow your results too much and end up missing some good content.
Note that words can change as well as just pop into existence. For example, one of the hot ideas of 2005, tagging, is not that much different from using keywords. But people call it “tagging,” and not “adding keywords.” So while I might have added "using keywords" to my searches at one point, I most certainly would have changed it to "tagging" in 2005.
Making your changes
I don’t think you should be hyper-vigilant about changing your queries—checking your keywords every day to make sure you’re on top of things. You’ll know when it’s time to remove things—when you notice that some of your queries are not getting much in the line of results, or that the results are getting worse and worse. Knowing when to add words to a query is trickier. I tend to add a word when I notice it’s in all my search results. If I’m not sure, I give it its own query and monitor it for a few days, and then integrate it into my other queries if I’m satisfied. I’ve narrowed my monitoring queries even further, which is always a good thing, and also updated them with new and relevant words.
You will learn a lot about new words and new things to add just from looking closely at what you’re already capturing. But I also recommend that you create a routine for yourself so that you can keep a more general ear to the ground and learn about new topics you might otherwise miss. As a special bonus, doing this will also clue you in on new sources to monitor.
An Ear to the Ground, a Nose to the Wind
Every time you research all the potential traps for a set of queries, you’ll come across blogs, or portals, or information sites, that cover your topic but do not get as narrowed or focused as you need them to. I recommend you pick a few of the best ones—not all of them, just a few—and read them regularly, either via Web page monitor or RSS feed. This does two things for you: it keeps you up to date on new information sources and teaches you new concepts.
The queries and sources you monitor can keep you up to date on new information sources that are relevant to your content, but your queries may be too focused for that to work well. You may get all kinds of sites about antique woodworking, for example, but nothing about general woodworking directories, which may be useful for you to monitor for antique woodworking sites. Use your queries to be as specific as possible, and then do a little general overview monitoring to hit all the topical stuff that can help you add new sources down the road.
Monitoring topical overview sites and news sources is also useful because these types of sites track all the trends for the topic you’re monitoring, and because they’re general sources, sometimes they will expose you to and teach you about a new concept or idea faster than if you had relied only on your queries. Which means that you can start experimenting with those concepts in your queries sooner rather than later.
It’s unlikely that you’ll have a problem finding topical Web sites that match your interest as you work on generating your specific monitoring queries. If you do, however, try a searchable subject index like Yahoo or the Open Directory Project.
Reevaluating Your Sources and Pruning Your Traps
I’m being very glib, aren’t I, recommending that you add queries when you might be feeling like you’re quite overloaded as it is? It’s just as important to remove queries that are no longer productive, or which aren’t quite what you’re looking for, as it is to add new names and concepts as you find them.
Sometimes you’ll discover that two sources you’re monitoring are overlapping more than a reasonable amount (perhaps five to ten percent), so one of them has to go. Or you may find that a source has become so overwhelmed with advertising that the difficulty of getting to the news outweighs the quality of the news. Time to get rid of it. You may also find that a site is almost never updating anymore, or has gone off in a wildly different direction. Get rid of it. If it comes back, you’ll hear about it from your general topic sources.
Here are some other signs that you may need to get rid of a source or a query:
• You never actively use the information that your trap finds. (This may also be a symptom that you have too many queries and are getting overwhelmed with information. Evaluate your sources.)
• By the time you get a good story from that source or query, you’ve already gotten it several times from other sources.
• When you get something from a source or query and you think, “Oh, argh.” Go with your gut instinct and get rid of it.
A Dynamically Changing Activity
If there were only one thing that I could really get across to you about the whole idea of information trapping, it would be this: it’s not a static process. You will constantly need to evaluate your results—not merely as informational items, but as elements of your monitoring strategy.
By shifting your focus from wandering on the Internet to setting traps for the information you want, you now can save a lot of time and effort in keeping abreast of the topics that interest you. I grant you, it’s a lot more time consuming to set the traps up than it is to just fire up your browser and start surfing. But you will more than make that time back in the long run.