“Lazy” Sourcing

booleanstringsBoolean, Diversity Leave a Comment


I imagine that you would agree with me that advanced Boolean searching for talent on Google is harder than searching on a job board or on LinkedIn, which, in turn, is harder than sorting and filtering a list of professionals in an Excel table. Sure enough, advanced search offers control over the results, but in some cases it becomes very elaborate even for skilled Sourcers – and inefficient as well.

An example is searching for “all” women names in searches for diversity. The challenges would be: limits on the length of the search string and the number of keywords; (severe) limits on the numbers of results displayed for any search; and some names that can be either male and female, to name a few.

As another example, an exhaustive Boolean search using job titles would require significant upfront research for what these professionals are called at target companies (that could vary greatly!) and will run into the search limitations as well.

Yet even if there’s a large Excel file, and even if some records have no relevance to the target whatsoever, you might pick up the promising records quickly. More generally, if you put your results into a system capable of searching, sorting, and filtering, that would make a difference in searching efficiency and the results. Searching in a set of records is much easier than searching among volumes of unstructured web pages.

So here’s a concept of  “Lazy Sourcing”. Post a job… no, I didn’t mean to talk about that.

“Lazy Sourcing”: get tons of info, perhaps most if it being irrelevant, then filter out what’s good.

As one example, you could search for text and Word files, that are, potentially, resumes, using simple, open-ended searching (vs. exhaustive keyword combination searching) and save them on your hard drive using Outwit Doc; then, search within the files. If many of the found files are not even resumes, that is fine too and is easier to digest when you have the files handy. (You can do the same with PDF files, but then you might need additional software for searching within the set – actually, if you have one to recommend, please do.)  Use a resume parser if you have access to one.

Another example would be collecting email addresses on an Association site (if it’s doable, of course), then cross-referencing against LinkedIn (just use the technique described in the referenced post) and only reviewing the records matching your target locations and companies. Note that this would not require any keyword searching at all. Additionally, if you did search with keywords you wouldn’t be finding many of the records that show us using this technique.

You would certainly need to eyeball the results before taking any action, but that is necessary no matter how you begin.






Leave a Reply

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