Nowadays, too many platforms and tools are claiming big data or AI-driven recruiting. There is a common pitfall here: a data scientist is not thinking in a recruiter’s way. How could we expect them to solve recruiting problems then? Before data scientists start to truly understand recruiters, don’t expect that big data can bring any miracles to your recruiting. Big-data driven recruiting can be fake unless data scientists truly “get you.”
As a data scientist, you can consider yourself an expert. No problem! Just acknowledge that you are not smarter than recruiters in recruiting. Listen to recruiters before you start to do your magic. Start with a conversation. You will find that recruiters are really nice people to talk to and that they have a lot of stories.
The conflict between data scientists and recruiters, on paper, is actually pretty easy to understand. We may have received very different degrees and were taught to think in very different ways. Recruiters look at the world as a big pool of people. Talent, skill, experience — these are the ways we categorize everyone we see. Words pour past our eyes as we scan thousands of emails and resumes looking for key phrases and fulfilled requirements. In contrast, if I’m understanding you right, data people look of the world and see not just people but statistical models, averages, norms, and outliers — what is the same, and what is not? Where does this person fit or why not? Part this, part here, part that, part where.
So to find a happy medium and a place to connect, think of the places where you overlap. Recruiters need to process huge amounts of information — that’s what data is all about! Finding quick ways to categorize things and understand people and experiences — data people already sees the world in that way. They find a way to bring it all together accurately, without losing what is really important, and constantly improve the way they do it.
Start that conversation, and begin to process those numbers a little differently. You’ll both benefit from it.
Big data must be valuable data before it starts to be powerful. Consider the problem of hiring a Principal iOS Engineer, from Mountain View, California, who is an expert in social networking, reputation management, and additionally, a bona fide basketball fan (a real challenge that our company faced as our client was one of the biggest NBA stars today).
There are many ways to approach this using data. You could find everyone who was an iOS Engineer to start with, and maybe quantify their profiles with a score that considers all the qualities you were looking for. The problem is, recruiters cannot score each candidate with a grade that says “fits” or not. They have to analyze the whole picture — iOS experience is important, but people’s years in the field are indicative of how good they are. We once had someone who claimed to have Swift, iOS’ programming language, experience for six years. If that had been true at the time, unless he was the architect and creator of Swift, he was a time traveler, because Swift has only been around for two years since 2014.
Then they have to compare their social media and reputation management rating, as well as their references and evidence of their talent, and weigh in what they know of their basketball fandom. And then factor in the current location or willingness to move to consider how “Mountain View, CA” they are.
While this can all be quantified in data, it’s not a black-and-white score or consideration. As a data scientist, these are ways you need to be aggregating and weighting your data to really be of service, to really create valuable data, for a recruiter.
So often a software or a tool overclaimed its capability. A good example is when big-data driven platforms are just a data aggregator of social networks and not much more. While these are certainly helpful, they are a far cry from the ultimate recruiting solution. Everyone claims to have solved recruiter’s problems but so far, they really have not.
Big data should be not only “BIG,” but also “smart.” Seventy-two percent of companies say that candidate quality is very important or critical in evaluating the overall quality of talent acquisition efforts.” To know a talent’s social profiles like their Linkedin, Facebook, or Twitter pages, or even getting their contact information, will not solve your recruiting problems — because recruiting at it’s best is not more contacts, it’s the best contacts.
The way to manage this is to examine the remaining problems that recruiters have. Your data should encompass their major issues without leading them down too many rabbit trails. Profile information, for instance, should be adaptive. While a Facebook or Twitter profile may be perfect for determining social media superpowers, github or bitbucket profiles are a much better assessor of coding and engineering abilities.
These days, I am thinking about this question frequently: what is truly missing in the recruiting industry? I am challenged by many of my friends who are experienced recruiters with the same question, too.
A database?
Do we need a better talent database? With LinkedIn, Connectifier, etc. we can search almost every employee in the market. It’s hard to make better database if the current talent database can support sourcing and recruiting well.
A platform?
If the database is large and completed enough, do we need better channels to reach out to talent? Do we need a better platform?
No. LinkedIn is the dominating social platform for professionals. We can freely connect to and reach out to talent. Beside LinkedIn, we have Facebook, Twitter, etc. for being connected with others. Furthermore, email and phone number can be easily bought from a lot of channels. A platform is not truly urgent need of the job market. After all, a professional platform is not easy to build up at all.
A management system?
There are over 128 ATSs in the market. The problem is which one to choose, as well as the effort it takes to for recruiters to use it.
What’s truly missing then?
There are a bunch of software tools and platforms to help recruiters hire fast, and Dean Da Costa has them all. However, there still remains the stupid question: “do recruiters still have pain?”
My personal opinion, evidenced by my time in the business, is that the biggest burden is in finding the right person quickly who is also interested in what you’re offering. Neither by itself is enough, which is what many of these platforms strive to do. The fact that there are millions of job seekers available does not make them fit for your position. The idea that you’ve found the next Steve Jobs with all the qualifications does not mean he’ll want to leave a cushy job at Apple for your low-paying startup who needs someone by yesterday.
Data scientists and recruiters of the world, unite! There are so many ways you can be more understood and useful to each other. Start with the conversation, and build your data and tools accordingly. Understand each other and build something fantastic together.
All in all, let us all turn back to the start point — understanding recruiters and recruiting.
I would like to hear more voices from our recruiting community. Am I right? What do you want that is missing? Is it better tools or better training or complete operations overhaul? What is your interpretation?
If you enjoyed or found this article useful please recommend and share. Let me know your thoughts and leave your comments below please if you have something interesting to share.
