Data Driven Recruitment And Intelligent Algorithms
The age when recruitment was limited to analyzing a resume, the accompanying cover letter, and scheduling one or more face-to-face interviews is gone.
Recruiters now can put to work algorithms parsing through vast amounts of data to gain a better understanding of the candidates queuing up in front of the building.
The aim is to sketch a very accurate professional and personal profile and find the best match possible, one that fits a firm’s corporate culture and projected public image.
Long-term studies have shown that skills and expertise are no longer the most important criteria that make up a viable candidate. Whether they like or not, recruiters had to embrace a holistic approach to the individuals passing through, paying attention to personality traits and other details that not long ago were dismissed as irrelevant for a job application.
There is lots of data on people, which can be procured within any organization and even externally – without hiring a private detective. Social Media channels, like Facebook or LinkedIn, or any other platforms that require creating a personal profile can offer a vast amount of information on personal preferences, social circles, professional connections, and even consumer habits, data that would be too cumbersome to ask during an interview. An applicant’s entire online presence surfaces one way or another!
Data mining like this, is currently being used for a variety of purposes, including recruitment, and the term Big Data has become synonymous with “improved” decision-making. Yet, the processing part is the key to success. Without clever methods and algorithms, Big Data would be exactly what its name first suggests – a vast, featureless collection of redundant information that is not only hard to read – but is impossible to translate into meaningful points.
Data driven recruitment does not always start with sophisticated black box algorithms, like the ones Google is using to rank pages in its search engine. A simple keyword filtering, although crude in nature, might narrow down the list of candidates. The simplest of software can scan resumes and cover letters for the words or phrases that might stimulate interest. Skimming through databases, can reveal individuals that applied earlier or who may have acted as clients in the past.
Intelligent Algorithms [IA] really step up the game when it comes to testing the candidates, and are especially handy for assessing soft skills, those that would be otherwise difficult to quantify.
Personality tests have been in the recruitment arsenal for quite some time, but their potential is enhanced by employing algorithms capable of identifying patterns and making predictions. Such an analysis can compare candidates with employees internally, so the organization can embrace an expandable metrics culture.
That brings us to the less explored – dark side of data driven recruitment – and the dangers of running a company with such a focus.
Modern recruiting processes try to reduce subjective (biase) factors, but may end up neglecting one important rule – collecting bits of information for any operation can create a bureaucratic mess that over-complicates the set of internal rules and regulations, and will turn an employee into a guinea pig – with sensors (attached) all over the workplace. Remember that if you get hired/reruited via a data driven recruitment method, you will step in your new office with those sensors already in place!
Recruiting via intelligent algorithms will always render a correct solution according to the set requirements. But, even the most refined analyses have a degree of uncertainty, largely because some of the variables may be dismissed as irrelevant. Historical performance is not a guarantee of future performance! And a candidate that passes all the selection filters – may not be a sure fit for the company’s long term goals.
Statistics performed on big data can point out the most favorable outcome, yet they often do that by excluding extremities, irregularities, and high levels of variance. The danger is obvious (think of the so-called fake war on talent) and the problems that rely on Technology for an answer. Relying on automation without understanding the mechanism underlying it, scary!
Data driven recruitment is currently replacing the old approach, one that relied on traditional metrics. The efficiency and quality of the hiring process tended to focus too much on itself, creating a loop that rarely rendered the most viable candidate. Parameters like time-to-hire (number of weeks to hire someone), cost-per-hire, and the size-of-the-shortlist created a false sense that the right candidate decision was being made.
The way we collect and use data has improved the talent acquisition world and data driven recruitment is making many processes easier and is more likely show the required results. Employers can certainly take advantage of data driven recruitment to make optimal decisions and expand their team – as long as it based on a fair IA process.
© New To HR