Advertisement

The Boolean Five All-Stars: Fame, Fortune, and a Free Happy Meal

Article main image
Aug 16, 2011

OK, I’ll admit it.

Promising fame, fortune, and a free Happy Meal was a ridiculous exaggeration. So to set the record straight, the free Happy Meal is off the table… unless of course you care to pay for it and eat at the counter.

What I can promise is that the Boolean Five All-Star pointed tips (not archery) below will, in a nutshell, give you the basics in how to build winning Boolean search strings.

This is not a book on the subject. These five points will just be a quick reference guide. The ‘book’ will be in a book I’m currently writing that’s been working in my head for the past ten years. More on that later.

Here are the five pointed tips:

  1. Set up your sets of parentheses connected with AND’s.
  2. (      ) AND (      ) AND (      )

  3. Enter a keyword into each of the sets of parentheses. For a wider search you can enter more than one keyword into each set of parentheses, with each of the keywords connected with an OR.
  4. The Boolean rule says that for you to surface a resume, one or more keywords from EACH set of parentheses much be found somewhere in the resume.
  5. Keep all RELATED OR EQUIVALENT keywords that are OR’ed together ALL WITHIN A SINGLE SET OF PARENTHESES.
  6. When selecting keywords in your search, for every keyword you select, ask yourself the following two questions:
    1. Will this keyword have a high probability of being found somewhere on most of the resumes I’m targeting? and
    2. At the same time, will this same selected keyword also have a high probability of NOT being on the resumes I’m not targeting, thereby screening them out?

Those are the five concise bullet points. Now we’ll take each one of the five points and add a little more explanation to each of them.

1. Set up your sets of parentheses connected with AND’s.

(      ) AND (      )

(      ) AND (      ) AND (      )

(      ) AND (      ) AND (      ) AND (      )

…etc.

You can have as many sets of parentheses / AND’s as you like, but the more sets of parentheses / AND’s you use, the fewer results you’ll get, but they’ll be more accurate. It will always be the subjective trade-off between low quantity/high accuracy and high quantity/low accuracy resumes.

In this Boolean world there are a lot of small quirks, one of which is the use of AND. If you write Human AND Resources, you’ll get the same results as if you had simply written Human Resources, and the AND would be understood to be there.

If you used parentheses and wrote (Human) AND (Resources), this would be the same as writing both Human Resources or Human AND Resources. When using single words like this, the parentheses add nothing.

It’s like 3 x 4 = 12 and (3 x 4) = 12.

Above is shown three different ways to write a Boolean search string for Human Resources. This Boolean search string would find all resumes that had Human Resources anywhere on the resume.

This search string of two words would also find resumes that I didn’t want because they were not Human Resources resumes. An example on an unwanted resume being surfaced would be one that somewhere in it had the following sentence:

the human mind provides many wonderful resources.

To get what we were really looking for we would need to add quotation marks:

“human resources”

Adding quotation marks lets us only look for the exact phrase.

What we have done here is a) think of words (human resources) that will include resumes we’re looking for and b) by adding quotation marks (“human resources”) we have enhanced the function of these words so that in addition to ‘including’ they will at the same time exclude resumes we’re not looking for.

So adding the quotation marks will exclude all resumes that have the following statement somewhere in them:

“the resources of a library boggle the human mind.”

In point 5 below, we’ll discuss how we can also use words to accomplish this same type of include/exclude function.

2. Enter a keyword into each of the sets of parentheses. For a wider search you can enter more than one keyword into each set of parentheses, with each of the keywords connected with an OR.

If we were looking for and Recruiter with IT experience, we could write the following Boolean search string:

(Recruiter) AND (IT) [I know, I know, at this point, the parentheses or the AND are not necessary but hang with me.]

But this string would miss all resumes that only had on them the following words related to recruiting: recruitment, recruiting, staffing specialist, talent acquisition specialist, etc.

To solve this problem and to make this into a wider search that would include resumes with any of the above words, we would write:

(recruiter OR recruitment OR “staffing specialist” OR “talent acquisition specialist”) AND IT

Good string so far, but as is, we would miss all resumes that didn’t have IT on them, but instead had Information Technology on them. So to cover this base, we would re-write the string as follows:

(recruiter OR recruitment OR “staffing specialist” OR “talent acquisition specialist”) AND (IT OR “Information Technology”)

The Parentheses Rule

Whenever you write a string that has both an OR and a AND in it, you’ve got to be using parentheses somewhere in the string. All examples below use the exact same numbers, 3,4,5,6 in the exact same positions; the same number and types of addition and multiplication signs; and the same number of parentheses, but look at all the different results you get by just moving the parentheses around:

(3 x 4) + (5 x 6) = 42

(3) x (4 + 5 x 6) = 102

(3) x (4 + 5) x (6) = 162

When you don’t put your parentheses in the proper positions, you’ll get undesirable results, but if you start off following the instructions in point #1 above, you’ll be heading in the right direction.

3. The Boolean rule says that for you to surface a resume, one or more keywords from EACH set of parentheses much be found somewhere in the resume.

This one is self-explanatory from the above examples.

4. Keep all RELATED OR EQUIVALENT keywords that are OR’ed together — such as (HR OR “Human Resources” OR “Talent Acquisition”) — ALL WITHIN A SINGLE SET OF PARENTHESES.

Suppose we had two family reunions — one for the H.R. Jones Family and the I.T. Williams Family. Everyone within each family reunion was related to each other. Each of them held their family reunion within one set of parentheses as shown below:

<H.R. Jones Family Reunion>

(HR OR “Human Resources” OR OD OR “Organizational Development”) AND

<I.T. Williams Family Reunion>

(IT OR “Information Technology”)

Everybody at each family reunion is related to one another and everybody is having a great time with BBQ chicken, mashed potatoes, biscuits, and good fellowship.

Then there was a falling out at the H.R. Jones reunion. OD left the family reunion because no one tried her lemon pie and she went over and joined the I.T. Williams reunion. IT got upset because no one wanted to listen to him sing, so he went to the H.R. Jones reunion. So below is what the family reunions look like now:

<H.R. Jones Reunion>

(HR OR IT OR “Human Resources” OR “Organizational Development”) AND

<I.T. Williams Reunion>

(OD OR “Information Technology”)

Remember the Boolean rule that says every resume pulled out of a database using a Boolean search string requires that every resume MUST HAVE ONE OR MORE WORDS ON IT FROM EACH SET OF PARENTHESES and we’re looking for a person that must have HR and IT experience.

Your search will indeed bring up resumes that have HR and IT experience on them, but…..

Your search will also bring up resumes that have HR and OD on them with no IT or Information Technology anywhere on them.

Your search will also bring up resumes that have IT or Information Technology on them and no HR or OD anywhere on them.

You won’t be happy with your results from this search, because all-star point # 4 above was not followed which says to keep all related or equivalent keywords contained within one set of parentheses. For happy results, KEEP IT ALL IN THE FAMILY!

5. When selecting keywords in your search, for every keyword you select, ask yourself the following two questions:

  1. Will this keyword have a high probability of being found somewhere on most of the resumes I’m targeting? and
  2. At the same time, will this same selected keyword also have a high probability of NOT being on the resumes I’m not targeting, thereby screening them out?

To start this section, I’ll repeat what was written above in Pointed Tip #1:

What we have done is a) think of words (human resources) that will include resumes we’re looking for and b) by adding quotation marks (“human resources”) we have changed the look of these words by adding quotation marks so that they will at the same time exclude resumes we’re not looking for.

To make it simple, the meeting of the “a)” condition in Pointed Tip #1 will, going forward, be referred to as the Include Condition and the meeting of the “b)” condition will be referred to as the Exclude Condition.

Another example would be if we were searching for a Java Software Engineer. If we chose Engineer as the keyword, it would do only a fair job of meeting the Include Condition because it would not include Java Software Developers and it would also only do a fair job of meeting the Exclude Condition because it would not exclude Electrical, Mechanical, and Railroad Engineers.

Using “Software Engineer” would be better because it would exclude Electrical, Mechanical, and Railroad Engineers but it would not exclude a very wide variety of Engineers that are not Java Software Engineers.

Java would be the best keyword in this situation because it would include all Java engineers and it would exclude all the other many types of software engineers. However it would not exclude those who stated in their resume that they would like a good hot cup of java to get them going in the morning.

This part of the discussion could go continue for a long time while going through all the possible combinations of keywords for this or for forming any other Boolean search string. At some point you have to come to terms with the realization that there is no one perfect way to write a good search string.

Personally I typically go through 3-5 iterations in trying a string, tweaking it, and then trying again. It is very much an iterative process, and instead of trying to reach perfection, find a string that yields what you consider a good amount of resumes for a given req and run with that. If none of them work out, you can always start the process again.

There are two excerpts from an earlier SourceCon article that will give more illustrations of the above five points. The article was entitled “Boolean in Disguise: LinkedIn Recruiter’s Cure for the Common Boolean Blues.” Within that article for Excerpt #1 is the following subheading: “Looking for a Used Car?” For Excerpt #2, look for the paragraph below and read down from there: ….

…these Boolean rules state that for a resume to be found, it has to have one or more Key Words from EACH set of parentheses. In this string we have four sets of parentheses:

(HR) AND (“Information Technology”) AND (Director) AND (“New York”)

The Book on Boolean

Again, the book on Boolean is a work in progress, but it is coming along.

In the article above, I’ve just given you the bare essentials, which are only the tip of the iceberg. To really teach Boolean recruiting adequately, my book will start very simply and slowly and builds from there, all the while giving practical examples each and every step along the way.

We’ll start with making blueberry pies, third grade arithmetic, family reunions, and examples of what happens when going through security checkpoints at an airport before you can board your plane. There’ll even be an example of a person making it through three sequencial interview rooms with interviewers in each room asking the candidate difference questions. To be hired,  the candidate must correctly answer questions from all three sets of interviewers before they can get hired. (This is one you’ll just have to see to understand, but trust me, for those confused by Boolean, this is one of the most effective ways to for you to understand and make Boolean work for you.)

From there we’ll gradually move into more technical requisitions that will touch on IT, software engineers, hardware designers, microprocessors, wireless, telecom, and other requisitions for other technology areas.

The 99% Factor

I started using Boolean searches extensively around 2000 and during the following eight years,

99% of these candidates came directly out of Monster using Boolean search techniques.

Since 2008, I have mostly used LinkedIn Recruiter (the monthly paid version — not the free version). Monster is still good, but now 99% of my candidates come out of the Boolean-searchable database of LinkedIn Recruiter.

The IRS Is Here to Help You

After the core Boolean section of the book, an IRS (Integrated Recruiting System) will be presented that takes the required skills which are turned into pre-qualifying questions for the candidates and finally they are turned into Yes/No questions for OFCCP compliance.

Tempted to Play the “Spam” Card?

Now, lest anyone be tempted to play the “spam” card, let me assure you that the Boolean strings I write are very specific and typically run 300-700 characters in length. This allows me to go into a database with the precision of a surgeon’s scalpel and extract just the type of resumes I’m seeking. In using LinkedIn Recruiter’s mass emailing function for the last three years, I have consistently maintained a Five-Star candidate feedback rating.

In Conclusion

It’s not that I was a super smart person to do all this, but instead I learned skills that enabled me to do well in technical recruiting. These are the skills I want to share with you, either in this book, a hands-on workshop working your own reqs, or a seminar.

If you have something you would like to see in the book that would clear up any aspect of Boolean strings searching or if you have any type of suggestion regarding Boolean training, please leave me a comment.

Get articles like this
in your inbox
Subscribe to our mailing list and get interesting articles about talent acquisition emailed weekly!
Advertisement