The Art of SEO Book The companion site to the book published by O'Reilly

Stephan

The following interview of Google expert Philipp Lenssen comes from my soon-to-be released ebook (2009 edition) on power searching with Google — available free to Art of SEO newsletter subscribers (sign-up is to the right on the sidebar). It’s over 50 pages and written from a market research and marketing perspective. No prerequisite knowledge of Google or SEO is required of the reader, but it’s no beginner’s text. I think even the advanced SEO will aget a lot of value out of it (or your money back!).

Philipp is a fellow O’Reilly author (Google Apps Hacks) and Google Blogoscoped‘s founder/blogger. Without any further ado…

Stephan: For what sort of research tasks is a major search engine not well suited?

Phillip: I think a search engine can be part of any research, with three caveats.

Added to these, there’s the challenge of being confronted with powers that be which may, every once in a while, work to make your research harder. It’s a more philosophical topic and not just a problem when using search engines. The basic communication approach of the powers is to hide, redefine, and distract. Let’s say John Doe is the leader of Doetonia, and he just got bribed with a big black suitcase filled with a billion Doetonia dollars. In utopia, informed citizens would be all over the place entering words like “john doe suitcase scandal” into the Google of their land. In the reality of our hypothetical example however, first of all, the scandal may not be known or if it’s known, it will be almost not covered (hiding!). Second, if the incident becomes known, the Doetonia press agency as well as other press houses which have an interest in the established system remaining stable can start to call it the “gentleman’s lapse of reason incident” or so, making it sound much more harmless — research this topic now using those words, and you may find some pros and cons in regards to the incident, but basically be stuck in a world defined by the Doe-establishment, because you’ve already started to use their words (redefining!). Third, if the incident becomes known and just reframing the issue won’t suffice, the Doe team and like-minded press can now create a red herring controversy in another area. Maybe they’ve just discovered that the pet cat of Ms. Doe has died in a car accident because there was no fence around the Doetonia leader house, and this becomes the discussion of the day, and all Doetonians jump to search engines to research the name of the cat, or whether the country should make fences mandatory for every citizen owning a pet cat (distraction!).

Stephan: For what sort of research tasks is Google not well suited, but another major search engine is? Please specify which search engine(s) you turn to in such occasions.

Philipp: TinEye.com as mentioned above is one. But Google itself got a lot of areas covered through its special search engines. I often go to Google Sets (labs.google.com/sets), for instance, because it lets me find new words for a given topic. You can enter e.g. “superman” and “batman”, and it will expand the list by returning “hulk”, “wonder woman” and the like. A sort of advanced version of this tool is becoming available with Google Squared (www.google.com/squared). These tools can be helpful if you’re still lacking the words to describe a concept, or if you don’t yet have a good overview of a certain area of knowledge.

One research tool I can highly recommend isn’t really a tool at all — it’s paid Q&A site Uclue.com. You set your price, ask a question, and have one of the researchers get back to you. The Uclue researchers were formerly working at Google Answers, before that got shut down.

Stephan: What are your favorite Google query operators, and why?

Philipp: Google has become fuzzier over the years, meaning they more frequently ignore certain words or automatically list results they consider to be targeting the word “you really meant.” But sometimes, you might have actually meant what you’ve originally written. In these cases, the plus (“+”) search operator comes in handy again.

One operator I use a lot is [site:example.com foobar], where example.com is the site you want to search across, and “foobar” is the keyword. You can even throw [intitle:something] into this mix to restrict it to sites from example.com which have the word “something” in their page title… perhaps because such pages are of a different type, and you only want to find that type. Say, [intitle:buy-this-product]. Note that the site operator also works with subdirectories, so [site:example.com/archive/2009/] is an option, too.

Stephan: Besides www.google.com, what are your favorite Google-owned websites, and why?

Philipp: I use Gmail a lot. When I’m in countries where YouTube is accessible, I use that a lot (though at the moment I’m living in China, where I can’t access it). Other Google tools like Google Image Search and Google Maps are very useful, too. Google Images expanded its feature set a lot during the last year or so. Now you can search by color, search for faces only and more. You can also use the site: operator, as described earlier, in conjunction with a Google Image search.

Stephan: What’s your web browser start page set to?

Philipp: It’s always set to a blank page, as is my desktop wallpaper. I like minimalist interfaces, they help me to focus. When I do want to search at Google, I can type “s keyword” in the Firefox address bar, which is connected to a Google search for that keyword. Or, quite often, I type “google” and hit Ctrl + Return, which will complete the domain name to http://www.google.com and let me start my search from there.

Stephan: What are your favorite third-party applications that are based on Google?

Philipp: Google’s search and translation APIs are nice. You can write interesting tools on top of that. I recently looked into writing a text editor based on JavaScript, which would run locally inside the browser but still be able to open files (you can google for “Netpadd B” to see what I came up with). In that plain text editor, you can mark any piece of text, hit a certain shortcut, and a translation of that text into English will pop up. (Or, if the text is English to begin with, it will be translated into Chinese.) There are many interesting tools and sites out there making use of Google’s APIs and gadgets.

Stephan: How/where did you learn so much about Google searching?

Philipp: I guess Google’s query syntax first and foremost targets casual users, so instead of half a dozen tuning options you can type in a casual query and have it return meaningful results. All the things for which that type of search doesn’t suffice, I usually learned from Google’s help files, blogs covering the subject, or from researching when writing for Google Blogoscoped (a blog and forum covering all things Google).

Stephan: Any training or resources (besides the ebook of course!) that you’d particularly recommend to anyone wanting to become an expert Google searcher?

Philipp: I would suggest reading blogs like Google Operating System or Google Blogoscoped, and also keep track of some of the official Google blogs.
You can pick up a lot of tricks this way.

Stephan: How does one assess the quality or credibility of the information produced by the search and various sources? Any practical tips beyond the obvious “buyer beware” type of advice?

Philipp: A huge number of details is involved. For instance:

Stephan: What one piece of advice about using Google as a research tool should the reader retain, if they remember nothing else?

Philipp: Sometimes, using a particular word may be necessary to find the right set of pages, even if that word may be synonymous to what you’ve already searched for. Try to imagine how an author of your imaginary ideal target page might phrase their sentences. Perhaps, when describing that baroque painting you vaguely remember and of which you forgot the artist name, you would need to write “voluptuous lady” instead of “overweight woman” to find the right result.