ATS Vendors

A Guide to Boolean Search Recruiting with Examples

Talent sourcing is simple. You have an open vacancy in your organization. All you have to do is perform search basis keywords, designation, location, job title, experience, skills and other requirements in your ATS which obviously will return the best profiles matching your query and behold! You have found your candidate.

Okay. That was all in good humor!

We all know talent sourcing is the trickiest part of hiring because the software that you rely on may return the highest qualified candidates, but it may or may not return the best candidates for that role depending on the search and matching capabilities it comes equipped with.  Today vendors may entice you with fancy words like Artificial Intelligence, Machine learning and cognitive computing but before these technologies exploded on the scene, most Applicant Tracking Software relied on Boolean search for their search and matching capabilities (and they still do).

Search engines like Google have been the flag bearer in this area as they first introduced Boolean in their search algorithms which was followed by much complex Semantic search and quickly progressed to AI based deep learning algorithms for returning search results to user queries. Recruitment has been quick to pick up the clues and has followed in the feet of the online search industry.  The current breed of recruitment software has vastly improved algorithms to source candidates more effectively.  We will see now how the search technology in ATS has moved from Boolean to Semantics and why the shift was necessary.

Boolean search strings in recruiting

Recruiters have relied on Boolean search for returning matching candidates from their company’s CV database in an ATS, or from other systems such as LinkedIn for a long time now. The search technology in most applicant tracking systems is an application of a standalone Boolean logic or combination of a two or more Boolean logics. Being equipped with a Boolean search functionality means you can use “OR”, “AND”, “NOT” , word operators to find candidates for particular requirement. In addition to these, we also have the “” and ().

So in total, Boolean logic works through five operators

  • OR
  • AND
  • NOT
  • “”
  • ()

For example, if you are searching for a marketing executive for your digital marketing team, and the manager insists on a Google Adwords certified individual, all you need to input is Marketing AND Google Adwords through a text query. If your recruiting solution has well-built taxonomy you might not even have to type Google Adwords in the query box, as it would automatically show in suggestions or fetch a Google Adword certified individual once you input Marketing as a key skill.  However, taxonomies are more of a semantic search concept than pure Boolean logic.

Other boolean search recruiting examples can be:

  • Job skill AND Experience AND Location

Software Development AND 5 years AND New York

A candidate who has software development as a skill,  5 years of experience and preferred location as New York.

  • (Job Skill OR Job Skill) AND Experience AND Location

(Software Development OR Software Testing) AND 5 years AND New York

A candidate who has software development or software testing (any one of the two) as skill and 5 years work experience and preferred location as New York

  • (Job Skill NOT Job Skill) AND (Location OR Location)

(Software Development NOT Software Testing) AND (New York OR London)

A candidate who has software development as a skill but software testing is excluded and preferred location is New York or London.

In an ATS these logics are also applied through check boxes and extensive filter

As it is evident, Boolean is a pretty easy way of constructing a query that fetches some relevant data.

Limitations of Boolean search techniques

True enough, boolean logic can find you candidates from your database as long as the search is syntactically correct. But that doesn’t mean it is finding you the best candidates. One limitation is that Boolean search works on exact terms specified in the query. For instance, a recruiter searching for “Web Designer” would certainly miss profiles that do not have the term and are alternatively listed as “Front End Designer”, or “UI Designer”.

The other problem that comes with Boolean search is the limited parameters around which you have design your query. For instance, if you need a candidate with 5 years’ experience in a particular skill set, such as software testing, there is hardly any way to specify this using Boolean, unless the developers of the Applicant Tracking System went to extreme pains of creating the most advanced Boolean search system. Another limitation is the sheer number of results that a Boolean query returns and that too not in any particular order. To get to the best possible candidate, the HR might have to open each profile and find the best match, which again is a time taking process depending on how large is your database of profiles.

Evolving to Semantic Search Recruiting

Some ATS’s have fought this drawback by providing functionality where candidates are divided in Premium and Other category based on predefined parameters like their College, Experience Years, and Current Employer. But that’s not good enough. Therefore, it’s time to look beyond and incorporate new search technologies with Boolean to make the most effective talent sourcing machine. With arrival of Semantic search and the age of artificial intelligence we have big hopes for sourcing and recruiting industry. How this search technology is different from their previous counterpart, we will study in our next post.

(Also Read: The Benefits of Semantic Search in Recruiting)

 

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *