Job Board Doctor Blog

Thinking beyond the search technology box: different tools for better results?

search technologyLast week Jakub Zavrel posted a fascinating piece on ERE.net that digs into search technology: Are Job Boards Still Relevant for the Future of Recruiting?  I strongly encourage you to read it (and definitely check out the comments – Steven Rothberg has some pertinent thoughts, too!).

In a nutshell, Zavrel discusses the role of search technology with regard to job boards’ longevity. Are two search boxes (one for title/skill, another for location) adequate? Is this the best that can be done? Or are there better alternatives? Not surprisingly, Zavrel – who runs a search technology company – believes that semantic search offers better results. He also believes that if job boards adopt this technology, they’ll be around for a while longer.

I’ll be completely upfront – I agree with him. Sure, the current search technology paradigm works, more or less, for certain groups of people – those who have been trained (even at some basic level) to know how to search. But did you notice all those qualifiers I threw in there? More or less, certain groups, etc.? Many times I’ve sat in a room with job search ‘newbies’ and watched them struggle with the current search model. How much information do they put in the skills box? How do they know which skills to put in? What does the location box mean – where they are now? Where they want to be? And what if the results aren’t what they wanted – then what do they do?

In the old days, people were actually taught how to search. I believe the discipline was called ‘information science’. The courses were typically aimed at those who would become the high priests of information – librarians, programmers, and researchers.

Guess what? It was that very formal discipline that led to the creation of our current search technology model. Yes, it can work. No, it isn’t always obvious or pretty. And yes, a lot of people out there get left in the dust.

So what?

Well, for job boards to maintain and expand their current position as a key method for employers and candidates to find each other, they need to innovate. That could mean integrating semantic search into their systems. That could mean aggregation of job content across multiple sources. That could mean peer review or crowdsourcing of candidates to provide more sophisticated profiles.

What’s the alternative? Well, job boards could just sit on their hands and hope things get better – magically.

Somehow, I don’t think that’s gonna work.

[Want to get Job Board Doctor posts via email? Subscribe here.].

[Check out the JobBoardGeek podcast archive!]

This Post Has 6 Comments

  1. Jeff,

    What exactly do you mean under “semantic search”?
    In particular, what semantic search technique should job boards implement first?

    I think that first step in improving quality of search results should be doing weighted keywords search that includes multiple keywords, so neither of keywords job seeker enters is mandatory.
    Would you consider that a semantic search?

  2. From Wikipedia: “Semantic search seeks to improve search accuracy by understanding searcher intent and the contextual meaning of terms as they appear in the searchable dataspace, whether on the Web or within a closed system, to generate more relevant results…Semantic search systems consider various points including context of search, location, intent, variation of words, synonyms, generalized and specialized queries, concept matching and natural language queries to provide relevant search results.” This is as accurate as anything I could come up with.

  3. Standard “keywords+location” search already covers most of that definition:
    1) Keywords define context and intent.
    2) Location field defines location.
    3) Synonyms dictionary in search engine do “variation of words”.
    What’s missing?

    What missing semantic search feature should job boards implement first?

  4. Dennis, I would have to disagree. Keywords don’t necessarily indicate intent, and many search systems are unable to handle variations of words (Marketing Dir or Marketing Director?) or natural language queries. A number of studies have shown that contextual/semantic search can produce more accurate results across a wider audience. At least from what I’ve seen.

  5. So what’s the alternative?

    “Semantic search” term is too broad.
    Do you have an example of semantic search technique that combines all these:
    1) Actually works.
    2) Feasible in implementation on a job board.
    3) Not about keywords and weighting them.

Comments are closed.

Back To Top
Search