Given the overwhelming, ever-exponentially-growing flood of information in the age of the Internet, being able to perform sophisticated searches is becoming so important that it isn’t too far-fetched to call it a literacy issue. Without these skills, you are in a sense in danger of becoming functionally illiterate in this brave new world.

Those individuals (and by extension businesses) with advanced search skills will be running circles around those without, because it saves so much time to search intelligently, and because a lot of answers can be found that are simply impossible to find otherwise. In a way, this separation into the search haves and have-nots has already been occurring over the last 5+ years.

And by the way, all of this isn’t simply about Google. Not at all. In a moment, I am going to walk you through a number of examples of advanced searches, and some of the tricks and techniques underlying them. But before I do, let me stress one other thing:

Even if you do only the most simple of "everyday" keyword searches, you are already going in the right direction. In fact, if you aren’t doing it already, make it a point for the next two weeks to stop yourself at every turn and ask: "Could I be doing a search right now to speed this up?"

I think you’ll find that the answer is almost always YES, and that it will be well worth your while to develop this as a new habit (a habit takes about 30 days of repetition to form).

Simply search for everything, and avoid using "manual" searching, i.e. avoid scrolling through documents, web pages, and lists both with your mouse and visually, asf. to find passages/names/etc. you’re looking for. Search options exist in Word, in your browser, on blogs, on Twitter, on Facebook, everywhere. Yet often we don’t use them, and the authors of software/Web tools don’t put sufficient front-and-center emphasis on search capabilities/ease-of-use.

For example, in your browser, never again manually search through long Blog comment threads or other large pages/articles manually, use your browser’s "Find" function and type the first few letters of your name or keyword, etc.

Granted, Gen-Yers on average are likely far ahead of all older generations when it comes to matter-of-cause use of Google, etc., however I doubt that even they know in large numbers about the kind of in depth, advanced search I am about to show you.

General Search Operator Considerations

Let’s first consider the most important search techniques by way of the so-called search operators. These may sometimes be accessible indirectly through a Web form under the heading of "Advanced Search", but originally they represent a kind of mini-programming language for telling the Search Engines what you want them to bring back. (Search Engines from here on shall include the "Search Function" in Web services other than stand-alone search engines.)

These are the "logical"/Boolean operators you may remember from math class or Logic 101 (fun, I know, but you really want to know a leetle bit about this, at least in these practical applications). Why know about these when you could also get most of the same results from using the Advanced Search forms?

Remember, this is about LITERACY. You want to become fluent in a secret language of sorts, and true command and mastery only come from truly delving into the heart of the matter. Plus, you will find that it is almost always faster to type queries into one search box than typing bits and pieces into Advanced Search forms which tend to look a little different for each service.

So let’s get started. I have made all of the examples clickable links, so that you can study the results. All results should be very similar on Google, Yahoo, and Microsoft’s Bing (formerly Live):

1) Nearly any search engine will assume by default that any separate words you type into the search box are meant as a logical AND, as in "show me all results matching BOTH this word-1 AND this word-2", though it may be in any order, and the words may be quite a distance from each other in the actual text.

You can usually place an AND operator without making a difference, e.g. for clarity in reading your search query, but mostly it will just look like this:

personal branding tips

2) To get a true phrase, a FIXED sequence of several words to match, you have to use ” ” (quotes) around the multi-word search term. Note that some search engines including Google will often bring a direct hit for a phrase to the top of the results heap, even if you didn’t use the quotes. But it’s not guaranteed, so using quoted phrases is much more precise, assuming that is what you are looking for. E.g.

"personal branding expert"

vs. personal branding expert

You can verify for yourself that this is more precise, by clicking both the quoted version and then the non-quoted one in Google, and comparing the number of results returned, in this case about 12,500 vs. 10 Million results (the count is in near the upper right corner in Google):

3) To get a logical OR (also called "inclusive OR"), as in "show me ALL the results matching this word-1 OR this word-2 OR this word-3", you simply type in "OR" between the keywords, or between keyword phrases in quotes:

"personal branding" OR "social branding" OR brand

Some search engines like FriendFeed’s Search also use a "," (comma) to represent an OR. (Either way, be sure to distinguish this OR from the so-called "Exclusive OR", which in essence says: Find only those results that have either Word-1 or Word-2, but not both". As far as I know, none of the search engines support this. Basically it would be like running to separate searches.)

4) Many search engines have an exclusion function using the "-" (dash/hyphen) operator followed by the keyword, phrase, or sometimes additional operator that you want excluded from the results. This in essence says: "Find all of the results for this word-1 except for those also containing word-2". E.g.

Yahoo: branding -"personal branding" -skin

Google: branding -"personal branding" -skin

would find all results containing branding, but not those also containing "personal branding", or those likely referring to skin branding instead of the marketing related kind. This would be a good search to narrow down results to those talking about corporate branding only (though you might find more exclusion terms to refine it even further).

By the way, there is typically no limit to the number of exclusions, though there may be a limit to the overall length of the query string you can submit to the search engine.

OK, with these preliminaries out of the way, let’s dig into the finer details of various key search engines or search functions on key services. Let’s start with Twitter, since it currently has the most buzz around its "Real-time Web Search" possibilities:

Twitter Search

Twitter Search is for now referring to search.twitter.com, as the Twitter Web interface integrated version is currently still somewhat limited/buggy in the result sets it returns. You are basically searching over every single public status update ("tweet") by any user, starting from the current moment and going backward over Twitter’s timeline. (If you are unfamiliar with Twitter or Twitter Search, read up on it here.)

Twitter Search allows all of the search operators already discussed, and additionally for the following:

1) "keyword(s) filter:links" – will seek out tweets containing the keyword or phrases and 1 or more links only. Nearly the same can be accomplished by searching for “http://”, though that will miss the few live links that Twitter recognizes from “www.domain.com/extension” type links.

It can largely be assumed that a tweet containing a link is more useful than one without, more likely to be chatter, unless the tweet is so sharp/witty/deep/inspirational that it would qualify as a quote (of course sometimes you may want to specifically look at the conversational chatter only – example of that further down):

"personal branding" filter:links

2) "from:username" and "to:username" – both of these can be very useful to query over your own tweetstream by topic/keyword, e.g. to find old tweets that you know you wrote, you know you wrote to somebody (containing certain link resources, etc.). Of course you can put any username you choose, and can therefore in principle back-trace all conversations between two users (each can only be used once in a given query).

You can also see if two users have been talking via Twitter’s so-called "@ replies" at all. If there’s no result returned, there was likely no direct communication, or at least recently:

from:unmarketing to:missive

As long as Twitter keeps back-data fully available in Search (currently, Twitter is unfortunately only letting you search back anywhere from 7 to 30 days depending on server loads), you could also use Twitter as a natural form of personal bookmarking this way. Nearly all of the “tags” are applied without extra work, simply as part of your tweets. A workaround to this problem of the backwards time limit is to also use FriendFeed and import your tweets there. FriendFeed currently places no such limitation. More below.

3) Searching for so-called hashtags – a keyword prefixed by "#" (pound sign) – is a way of detecting additional intentionality about tweets. Either it serves as a point of emphasis/visibility by the author (since a common keyword like "#branding" or "#quote" would still show up in a search results even without the specific # prefix), or more commonly, if the hashtag is a unique abbreviation, it serves as a sort of code to be specifically searched for by those that know about it.

This is most commonly done for conferences (recent examples are #140tc and #twtrcon), for ongoing weekly Twitter-based discussions around a given topic, e.g.

#journchat

is for journalism discussions on Monday evenings U.S. Time, or as a meme that becomes self-replicating enough that people participate, and the hashtag gets into the Top 10 "trending" keywords/phrases on Twitter for a while.

Either way, the authors of tweets using hashtags went to the trouble of using the # symbol and/or created a hashtag to highlight something. Use that knowledge to your advantage when searching.

4) "since:timestamp" and "until:timestamp" will allow you to segment out tweets from a specific day or number of days, as needed. This can be useful if you wanted to e.g. view only those tweets for a conference that were actually sent during the duration of the conference, and leave out the chatter before or after, which is e.g. less likely to contain "twitter-casting" of the actual conference panels.

#140conf since:2009-06-16 until:2009-06-17

5) "near:city-name" – this operator will find tweets that originated from a user account that Twitter thinks is the city name you are referring to. Since this is going off of users’ self-reported location field in their profile (and NOT off of some precise geo-tagging a la iPhone location, though Twitter is reportedly working on that), which is free text, and for some contains things like multiple cities, "everywhere", "The Interwebs", asf. this is not particularly precise, but it can still work in aggregate. E.g.

#140conf since:2009-06-16 until:2009-06-17 near:sf

will find all tweets about the TwtrCon Conference that were placed by users based out of San Francisco, though Twitter has no idea (yet) whether they were at the conference in New York or just talking about it.

5) To bring it all together, and for a special tip, we should also consider the so-called Retweet convention on Twitter, a format which allows one to quickly copy & paste a given (useful, funny, etc.) tweet from another user, and forward it on to our own Twitter network of followers, while giving credit to the original author. E.g. I tweeted

"RT @mvolpe: New Blog Post: Are Your Compelling Offers Actually..Compelling? http://tinyurl.com/mlbzw4"

giving credit to user @mvolpe, and used the "RT" prefix to signify the retweet. This is actually a convention that spontaneously arose from the user base (another format uses "via @username", used most often if the tweet text is sufficiently altered, but credit for the find is still meant to be conveyed).

What this means for our searches is that we can either search specifically for "RT OR via" to find tweets that were deemed worthy of retweeting (there are actually entire third-party services set up keeping track of these counts, and thereby surfacing tweets according to their presumed repetition popularity), or, we can exclude those tweets to avoid a lot of duplicates!

So here is a great way to cut down on overly large result sets, taking out most "link-less" chatter and Retweet duplications, as well as "psychology jobs" related postings:

psychology filter:links -RT -via -job -jobs

[As an aside, though still search literacy/awareness relevant:

I use this very example query above, and then pipe the RSS feed from the result into a somewhat more permanent receptacle such as FriendFeed or a Tumblr mini-blog. Remember, Twitter might (and currently does) cut off the backwards reach of your result sets, currently during heavy daytime loads it’s at most about 7 days back. This presents a real problem for your own research/archiving purposes.

Part of the reason may be that Twitter is thinking about making long-range backward data mining a “for pay” feature that large corporate marketing agencies, etc. may pay them a lot of money for (obviously not if they could access everything for free through Search.twitter.com). Only time will tell, though I think it is definitely important for the community to be aware of this possible issue.]

Or, here is another complex example to search for the term "mashable" while excluding tweets from the username "mashable", any @ mentions (or replies to) username "mashable", and tweets with links. Remember how I said earlier that you could do exclusions on some operators? This is an example of that:

mashable -from:mashable -"@mashable" -filter:links

This could be used so that you see what people are saying about Mashable, the blog, that is NOT one of the countless retweets of @mashable, not a tweet from "@mashable" himself, and doesn’t include links to further content. In other words, what people are saying about that brand the most raw and unvarnished form.

FriendFeed Search

OK, upon writing this section on FriendFeed power search, I realized that this post was getting to be really long. So rather than overload everyone, I figured I’d push this and the section on Google search tricks into a follow-up post in a few days.

I hope you found this enlightening, and that you take the time to practice advanced search. To become "fluent" and fully "search literate", you will need to practice. I know that saying this in our ADD world is somewhat of a bummer, but the payoff, especially for your business, can be tremendous. Remember, running circles around your competition and all of that…

If you’re not on Twitter yet, you should be, if only to see what’s going on, and to grab any usernames (for your own name, your company, and your products/brands) that may still be available before someone else does.

(If you are completely new to Twitter, first watch this brief video, and click though this presentation slide deck.)

Even if you decide that you don’t have the time to invest in maintaining an active profile on Twitter, you should at an absolute minimum understand that the new "real-time Web" that is emerging due to Twitter’s popularity is changing the game in many ways:

Not only is it causing redesign changes and opening-up at Twitter’s rival social media services such as FriendFeed and Facebook. Search of Twitter’s massive real-time stream of "Tweets" (the micro-messages that users send to their follower lists), is now being called "the pulse of this society" by wine merchant turned Social Media guru Gary Vaynerchuck. And I would agree:

Should you know what 10 14 Million people (yes, it grew by nearly 50% in the last month), many of them sought-after influencers and early-adopters are saying about you, your company, your brands, your products, your market, and your business’ target keywords on Twitter?

Of course you should. Twitter’s recent geometric growth proves that it is finding itself right at the inflection point from early to mainstream adoption in these last few months. And therefore searching at Seach.twitter.com has become an absolute goldmine of marketing relevant information, one that must almost be considered indispensable at this point:

So much so that there has been much rumor and speculation lately regarding a possible acquisition of Twitter by either Google or Microsoft. In Google’s case, it would be mainly a way to preempt Microsoft or other competitors from disrupting its search dominance. In fact, I have recently argued that acquiring Twitter may well be Microsoft’s last best hope at getting back into the search game in earnest.

This is because Twitter’s so-called Timeline makes its search results extremely, well, timely and time-bound, unlike Google’s and other search engine’s authority models that can feel stale in comparison to Twitter in many circumstances (and don’t allow for a clear sorting by recency of e.g. the 1,000 top results).

A good recent example were searches for SXSW Interactive Convention  panel attendee notes: The results via Search.twitter.com taken from Twitter’s Timeline are much fresher & more pertinent, especially during and right after the conference (one caveat is that Twitter will currently at times cut off backwards results beyond 7 days, apparently due to server load issues).

Google on the other hand will serve up mostly results from SXSWi 2008, 2007, asf. due to its authority model (the pages obviously had much longer to gather links from everywhere on the net, through blog post mentions, social bookmarking, etc.).

So again, one could easily see how Google simply cannot afford to let anyone else walk away with this whole new category/aspect of "real-time Web" search, which in itself has a strong component of recommendation filtering: Twitter users mostly only pass on those items and links that they feel would add value to their followers.

That’s because in a sense their reputation is on the line with every tweet they post. They know that almost instantaneously dozens, hundreds, or even thousands of people might hold them accountable if they send them garbage. Compare that to Google, where the ever-evolving practices of search engine optimization (SEO) have already greatly distorted the results that Google will spit out for most commercially relevant keywords.

In a way, Twitter is succeeding, to some extent by seeming accident, at crowd-sourced search models where the likes of Jason Kalakanis’ Mahalo and recently Wikia Search (started by Wikipedia’s Jimmy Wales) have failed, largely because they could never get enough users and user buy-in to scour enough content. On Twitter, content recommendation happens naturally, as a side-effect of the ongoing conversations or (more or less helpful) "stream of consciousness broadcasting" (sometimes referred to as "life casting").

And incidentally, each tweet containing a link to further content tends to also have a good amount of semantic/meta-data accompanying that link, without much of the nuisance of formal methods such as tagging, because the descriptors are added naturally as I already mentioned.

Are you beginning to see why Twitter is so special? Why you should start running searches at Search.twitter.com today if you haven’t done so up to this point? Go do it, right now. Go there and input queries for your most important keywords.

As gary Vaynerchuck says, (rough quote) "if your business is selling soup, you should go there and monitor the pulse of soup in society"…

(A follow-up post "10 Deep Points About Twitter" is on its way, stay tuned…)