I have tried to define how search works – in one page. In reality things are more complicated than this. However, I believe that understanding search mechanisms even a little bit helps searching.

Here you go:

Internet search engines:

• —  …crawl the web, going through links and memorizing the content of pages.
• —  A page has little in terms of its contents’ structure: a title, a URL, words, phrases, images.
• —  Search can be done in terms of keywords or phrases (KEYWORD=”engineer”).
• —  Pages are ranked. One of the factors in getting a higher rank is having many links point to the page.

ž

Databases:

• —  …are much smaller than the web.
• —  Examples: Job boards, LinkedIn, your ATS.
• —  Every record is structured and has predefined fields (as an example, name, title, company).
• —  A database is indexed, so that search can be done by the fields in its records (TITLE=“engineer”).
• ž  Some databases (such as LinkedIn) expose part of the content to the surface web.

žThe Boolean logic (AND, OR, NOT) allows to combine search terms and provide instructions on how to search on databases and on the web. Search capability and syntax are different in different places.

Display of results:

• Search engines at first show preliminary results and an estimated number of them.
• If we go through the pages of results and select “show all results, omitted included”, the engine finalizes the list of results.
• No engine will show more than 1,000 results.
• Databases can show all results (some, like LinkedIn, charge extra to show more).

Food for thought: If we get a limited number of results from the web, parse them and put them into our own database we can search again, in a more powerful manner.

For those who wanted to see the top 25 strings: I am afraid that this post will not be what you expect, but I hope it will be useful, so please read on!

Well, it’s not that I have changed my mind about sharing them. I love sharing my sourcing knowledge in posts, lectures, and presentations. When I do sourcing contracts I offer to explain how I get the results as part of my service. I have sourced for many different verticals recently, such as energy managers, chemistry PhD’s, software engineers from Google, telecom consultants, marketing managers, quality control specialists, safety system managers, and truck driver recruiters, to name a few. It is amazing that similar sourcing approaches work. After you’ve spent some time researching and figuring out the terminology, certifications, organizations, online gatherings, geography, and target companies, you can do similar things and find candidates.

However, that common sourcing approach cannot be based on particular search strings – which I have been trying to communicate to my students and readers for quite a while. There’s a belief that there are some magic strings that would provide results for all. If you go back to my article you will see one sentence in red that says it’s not true. It’s also mentioned in a post about sourcing myths. The post about my favorite strings is as close as I can offer if we are not talking about anything specific.

So in a sense I am going to disappoint you; though I am glad I got your attention. The top 5-10-25 strings are as unlikely to exist as the top 5 gadgets for cooking, the top 20 sentences to start a book, or the top 10 ways to build a piece of furniture.

It’s true that I often start with

• “intitle:resume OR inurl:resume” on Google, or
• “public profile powered by” site:linkedin.com on Yahoo/Bing, or
• filetype:xls name title company (adding keywords) on either Google or Bing,

but then I always change the strings, usually dozens of times. The search results guide me how to modify a string or whether to start at a totally different point. In particular, while “intitle:resume OR inurl:resume” may sound like a good start, for some job openings it doesn’t bring any results, while for other it may bring way too many. Also, as we know, the Boolean search syntax on Google, Bing, LinkedIn, Twitter and other places is different; so, many Google strings can’t be reused on Bing, and so on.

Sourcing is not rocket science. But there’s no magical one-size fits all cheat-sheets or sets of strings. Practicing, refining your strings, following the correct syntax for the site you are using, and reviewing the results will “get you there”, while sets of strings offered by others will not. Those “top” strings offered by others are just samples (if they follow correct syntax, of course – many don’t, unfortunately); you need your own.

If you are new to Boolean search, ask me about the beginner webinar I did recently; I offer the slides and the recording. Hanging out at the advanced search page on Google or Yahoo and going over help pages and examples would also make an excellent introduction to advance searching.

If you have a specific niche or a job description that you are working on and have questions, please feel free to post on our Boolean Strings Group and Network . If you have found interesting aspects of search syntax or interesting sourcing examples, please share with us. We have bi-weekly online chats on the Ning site; you are welcome to join for information exchange. Also, I am thinking of publishing (or showing at a webinar) some sets of strings that I use for a particular search (that would be lots of strings for any search!) with explanations of the logic. If you could do that for any of your searches, that would be great too.

As Christine McKenzie said it,

“While there isnt a top 10-25 list, there are consistent approaches to take that can be tailored.”

While sourcing is not just about search, the search is a critical part of it. I’d like to discuss the seven Best Practices around SEARCH as the key element of sourcing.

Here you go.

1. Research before you search. If you have a job opening and are looking for candidates, figure out the title synonyms, major keywords, target companies, geography, certifications, organizations, conferences, places where “these people” hang out, especially if the role is new to you.

2. Use the correct search syntax. Search on a computer may correct your spelling using a built-in dictionary. But, unlike us humans, it will not recognize other syntax mistakes. Seemingly minor things such as a missing quotation mark or an extra space after a Google operator will produce results that you are not looking for. Search for special characters such as @&%$! is not possible on Bing or Google.

3. Refine and vary the search strings. The search process is about getting a good amount of relevant information to process. Perfectionists want to get relevant results only but it slows them down. 🙂 If you have “the wrong” results, too few or too many results, it’s a great prompt for what to do next. Refine your strings, vary the sites and approaches, and keep collecting the data.

4.  Practice. No webinars, training DVDs, tutorials, built-in strings, or help pages will lead to success unless you practice the search skills from day one. There is no magical list of N strings that you can base your searches on. Pre-cooked Boolean strings are just samples; you need your own strings.

(Please note: As of 3/18 the list of the top 25 strings is no longer available.)

5. Search only for part of information. Some keywords like the state of OR are a bit hard to find. Qualities such as the degree or years of experience may be challenging to spell out in a search string. You will have a chance to screen the results for those qualities.

6. Automate the parsing, not the search. Search shortcuts, systems with hidden Boolean syntax, meta-search engines and systems that find resumes based on a job description (imagine that!) may sound good. However, if you use them you will be missing results. What is great to automate is the information collection, sorting and parsing.  I use browser add-ons to auto-scroll, parsers to extract files or contact information and MS Excel to sort and filter.

7. Cross-reference. We’ve long moved beyond just searching for resumes. There are profiles, lists with contact information, blogs and forums rich with traces of potential candidates. These candidates have “distributed profiles“. As an example, I often find people in Jigsaw and Zoominfo, then look them up on LinkedIn, pipl.com, Google and Bing.

Thanks for reading! Feedback and comment are very welcome.

Open Groups on LinkedIn is a great idea! (I have been long complaining how I wanted at least job posts on groups to be open to all.) Open groups add valuable content to the “surface”, searchable web.

There are several ways recruiters can benefit from the new feature. Here are a couple of suggestions.

1. When you post a job on an open group it’s given a static URL that you can then share on your LI status, FB, Twitter, – and also with individuals and on other groups. Just make sure that you “mark the item as a job” if you share on another group.

(Did you notice that sharing on LinkedIn now allows to enter email addresses, along with picking recipients from your list contacts? This is another fantastic, though somewhat buggy, feature.)

2. Note how the static URLs of the posts on open groups are generated. They pick words from the group name for groups and from the discussion titles for posts. Therefore, we can search for open groups like this, for example:

site:http://www.linkedin.com/groups/ inurl:boolean strings

and for discussions or job posts like this:

site:http://www.linkedin.com/groups/ inurl:recruiting send OR apply OR submit

You might want to create titles for your discussions and jobs that contain good keywords, since search engines give higher weights to words in URLs when ranking.

Most of us have target companies in mind when we search. For some of our jobs at Brain Gain Recruiting the target companies are the “Big 4” and other large consulting companies. LinkedIn conveniently offers us to search by past or by present companies. If you start typing a company name in the advanced people search dialog, it prompts for including companies as you type.

There are some subtleties though and I’d like to point out some of them today.

• How many Big 4 com companies do you think there are? Correct, those companies merged, bought each other and split businesses in the past. However, it’s not the end of our investigation. Companies on LinkedIn are entered and edited by employees. As a result, as an example, these companies are listed as different ones: Deloitte Consulting and Deloitte; PwC Consulting and PwC; Accenture and Accenture Technology Solutions. If we work with the company prompt in the people search dialog, we need to take this into account: look for and use all possible spellings of a company name.
• Looking this way we will still miss some people, because they entered their company name in an unusual format or with some additional info. This LinkedIn member lists his employers as “Accenture – Large Consumer Electronics Retailer“, “Accenture – Large Semi-Conductor Equipment Manufacturer“, and “Accenture – South African Mint“. He can be found by searching in the keywords field but not by picking the company from the menu list.
• Further, LinkedIn allows people to add new companies to their profiles and not close the date on the past company. This is reasonable, but it’s important not to miss a current target company. This profile (found by this string by the way) has 3 current companies listed.
• Finally, those of us with paid accounts have access to “total years of experience” or “years of experience at a company”. We need to keep in mind that poorly filled out profiles (which doesn’t mean poor candidates at all) will not show up in the right place in those searches; here’s an example.

Many of the same considerations can be successfully applied to other candidate search systems.

Hope that helps. 🙂

I am excited to be  a track leader and teach sourcing mastery at my second #trulondon event hosted by the wonderful Bill Boorman, the true leader of recruiting unconferences. The last #trulondon was, perhaps, the most content-rich event I had ever been to; was it in spite or because of the lack of preparations, presentations, and slides? The days were packed with interaction, ideas, technical tips, advice, and brainstorming. Katharine Robinson “the Sourceress” and I taught an all-day interactive sourcing class after just having met, and it was a happy experience. I am looking forward to representing sourcing knowledge and enthusiasm at #trulondon/#trusource again this year.

For the folks who haven’t met me: I work as a Partner at Brain Gain Recruiting, performing sourcing for our own contingency searches as well as for others (companies, agencies) on a contract basis. I used to be a Software Manager and an Engineer, and have a degree in Math. This helps me to understand hiring managers and to navigate web searching techniques. I moderate web sourcing communities called “Boolean Strings” on Ning and on LinkedIn. Here are my profiles: Irina on LinkedIn, @braingain on Twitter. My blog has some tips for sourcers.

Here’s what I can offer at the upcoming unconference:  hands-on sourcing for your particular requirements (bring them with you); Q&A; tips and tricks; and demos of cool sourcing tools.

Here are some questions I can answer. How does one get going with advanced searches on Google? How do Google and Bing compare? What are “people search engines”?  Will semantic search replace Boolean? How is LinkedIn different from a job board? How many people can you find on LinkedIn? Which account type is the best? How can you send multiple messages on LinkedIn in one click? Can you recruit on Twitter? How do you include a location in a search? How do sourcing techniques vary depending on the country/region? How do you search for a specific degree? What can you do if you have no results in a search? What can you do if the results are irrelevant? How do you speed up your search results’ filtering? How do you find out the contact information of a LinkedIn member? How do you find lists of people with contact info? How do you verify email addresses? What is the best way to initiate a contact with a prospect you’ve found? (and a few more).

Hope to see you in London on Feb 15-18, 2011!

Imagine that you could get the magical core list of search strings somewhere and just keep using them… The truth is, there isn’t one. Pre-built sets of strings are good as examples but that’s it. Every search is unique, and the info out there gets updates continuously, so you need your own strings. I was glad to hear at a recent joint webinar with Andrea Mitchell – from across the globe from me – that she shares this view with me!  🙂

That said, some of the same questions and examples come up on a regular basis at the Boolean Strings Group and Network , so I thought I’d list a few strings here.

In all cases, add variations of your keywords. Single words are always better than phrases, unless the phrase must be there, as in the case of the “SQL Server“, which is a product.

• Look for resumes on Google intitle:resume OR inurl:resume -job -jobs. There’s usually no need to put all sorts of synonyms for the word resume. Change “resume” for “CV” if you are in Europe or are looking for an academic. Looking for intitle:vitae when you are searching for a Java Engineer in the Bay Area is not likely to bring many results. Use filetype:PDF OR filetype:DOC, perhaps along with the words you’d find in a resume, to discover more.
• Look for LinkedIn profiles in the US on Google site:www.linkedin.com inurl:pub OR inurl:in-inurl:dir -inurl:jobs Add the exact location names as LinkedIn spells them. (X-raying is cool but searching within LinkedIn may be more productive because it is structured.) For the two-letter-abbreviated geographies such as uk.linkedin.com or au.linkedin.com searching is a little easier and may look like site:au.linkedin.com inurl:in OR inurl:pub -intitle:directory
• Yahoo/Bing seems to find more results and would happily pick many profiles with words like “view full profile” or “public profile powered by” or even just powered as opposed to Google. On Yahoo, search for powered site:www.linkedin.com. Don’t forget the parenthesis around OR statements on Yahoo/Bing.
• Find email pattens: search for email * * company.com on Google. Vary the number of asterisks. You may want to exclude words like info, support, or the company site and add words like my to get there faster.
• Looking for lists: filetype:xls name title company phone email -form. It helps to add the names of your target companies to this one.
• Here is an interesing advanced string: intitle:”index of” modified directory may find directories with files with all sorts of information.

Altering strings multiple times, examining the strings for correct syntax and the results for guidance on how to modify strings, and being creative always helps.

Your comments and strings are very welcome!

I am forming two Boolean Sourcing Mastery Classes, for people who would like to bring their Internet Search skills to a true Boolean Master’s level:

• Boolean Mastery Class Level I (Beginner-to-Intermediate level),
• Boolean Mastery Class Level II (Intermediate-to-Advanced level).

Each group will meet by webinar, twice monthly, 90 minutes each, for a total of twelve sessions. Sessions will start mid-January 2011 and last for six months.

Every session will have follow-up on the previous session, lecture, demos, and interactive discussion. I will provide hand-outs, homework, and quizzes on the material. At the end of the semester participants will get a DVD with all the materials.

If you sign up, please expect to actively participate in discussions, practice and apply what you learn.

I will be available for Q&A (and also, if necessary, to catch up if you miss a session), by email and by phone.

Expect your sourcing mastery, productivity, and rate of success to rise with every session.

Here’s the Program Outline:

• An Overview of Sourcing
• Boolean Syntax
• Major Search Engines
  • Google
  • Bing/Yahoo
  • Advanced Boolean Operators
• Industry/Terminology Search
• Geography Search
• LinkedIn
  • Advanced People Search
  • Tips and Tricks
• Twitter
• People Search Engines
• Cross-Referencing Information
• Specialized Search
  • Vertical Search Engines
  • Meta-search Engines
  • Semantic Search
  • Deep Web
• Productivity Tools
• [For the Advanced Group] Custom Search Engines
• [For the Advanced Group] Advanced Sourcing Shortcuts
• New Tools and Sites (as they come up :))

At the end of the program the participants who successfully pass the final quiz will be awarded the titles and receive certifications of a Boolean Master or a Boolean Advanced Master (for the two groups).

You will feel confident, comfortable with tools and terminology, will successfully apply new sourcing skills and knowledge to your practice, and – most importantly – have fun doing advanced online sourcing.

Sessions begin mid-January 2011. Price: $1,999 before 31/1/2010; $2,299 after 1/1/2010. (Compare to AIRS pricing.)

Payments need to be sent to: Irina Shamaeva, Brain Gain Recruiting, 201 Vista Heights Road, El Cerrito CA 94530.

Space is limited to a small number of participants, so that everybody gets personal attention.

Questions? Please call 510-233-9493 (Pacific time) or email me at  [email protected]

Thanks to heads-up from @SourceConNews I took a look at the Kryptos Challenge and spent a bit of time trying to figure it out. Challenges are addictive! 🙂

I have no knowledge of cryptography. But the author, Jim Sanborn, has just revealed a clue, and (who knows) maybe it’s not too hard now.

The way I see it, the K4 challenge has two sides. First is the coded 97-letter piece, that needs to be converted to plain English. Here it is:

OBKR

UOXOGHULBSOLIFBBWFLRVQQPRNGKSSO

TWTQSJQSSEKZZWATJKLUDIAWINFBNYP

VTTMZFPKWGDKZXTJCDIGKUHUAUEKCAR

The just-revealed piece of K4 is the word Berlin, which, when K4 is decoded, should appear in the positions 64-69.

Second, there’s the linguistic (or the semantic?) side of things, that should help decide what this last piece is about. While I have made no progress on decoding whatsoever, I am pretty sure I know what the text, that is to be discovered in K4, will say. Let me share it.

To start with, here are some hints coming from the parts K1-K3 and also the Morse Code that can be found near the sculpture (check the links above for more details on #Kryptos):

• (K1) “…lies the nuance of iqlusion” Misspelled (i.e. important) word illusion
• (Morse) “Shadow forces“
• (Morse) “Virtually invisible“
• (K1) “Absence of light“
• (K2) “The information was gathered and transmitted undergruund“. Misspelled (i.e. important) word underground
• (K3) The question “Can you see anything?” related to the discovery of the King Tutankhamun’s Tomb in 1922 – to which the answer was, “Yes, wonderful things”, and there was one other chamber discovered after the first, with lots of golden things (and then a third one, with the body of the King)
• (Jim’s hint) Directions; and (K2) North. West
• Pond with ducks near the sculpture

OK, and here is what this all relates to. Compare the highlighted words above with the highlighted words below. (And it’s not the Berlin Wall, as most posts seem to point to!)

The Berlin Tunnel was a project code-named Operation Gold. The tunnel was build by the CIA together with the British Secret Intelligence Service, to tap conversations on the East side of Berlin. It was a large information gathering project.

There was illusion on both sides. The Americans felt that the project was successful, while the Russians knew about it from day one from a British double agent (a mole) George Blake. The fact of the spy’s actions became known about five years after the Tunnel was “discovered” by the Russians. The story is even more complicated, as the KGB, after learning about the tunnel from Mr. Blake, did not let the Soviet Army know, so it’s likely that conversations that were tapped were between people who had no idea about being heard. (It’s quite a story, I recommend reading about it!).

I am wondering whether the sculpture itself is supposed to remind us of the iron curtain.

Also, take a look at the article: BERLIN: Wonderful Tunnel (1956). It gives us a few more initial details on the Tunnel discovery, and its style seems more like “Jim’s style” in k1-k3 to me.

“…Self-important ducks and chickens strut like commissars in Alt-Glienicke’s cobbled streets…Mobile generators were humming to provide lights for the occasion, and at the entrance to a hole dug in the ground, a colonel of the Russian signal corps was on hand to explain it all. Ten feet below, its entrance a hole cut in the roof by the Russians, lay the tunnel itself: a cast-iron tube about six feet in diameter and 500-600 yards long, crammed with electronic equipment, cables, tape recorders, ventilating apparatus and pumps of both British and American make. At the East German end, cables led out of the main body of the tunnel to a separate chamber where they were linked to two East German cables and a third used by the Russians. What was at the American end? The newsmen were not permitted to know…”