Recruitment has traditionally been a somewhat hit-and-miss affair. In fact if statistics are to be believed, almost half of appointments end up as failures within 18 months.
That shouldn’t be surprising – personalities are often the hardest things to accurately convey and interpret in an interview, and 89% of these failures are blamed on “attitudinal reasons”.
It really should be more surprising that many companies still make vital hiring decisions – with all the financial liabilities that they entail – based on an often-brief and unstructured face-to-face interview and a cursory review of the information provided by the applicant in their resume.
Unfortunately much of the time the outcome is still dictated by an interviewer’s “gut feeling” – that often deceptive sixth-sense, quite common among people in positions of responsibility in business, that their own instincts are good enough to base important decisions on.
Going Beyond Gut Feeling
It looks as if those days are coming to an end. Data (and especially Big Data) can help take the guesswork out of recruiting just as it can with marketing and manufacturing processes.
Technology exists today which allows recruiters to compare the personal data on your CV with hundreds or thousands of other applicants and employees – in similar positions at their business or others’ – to get an idea of how long you are likely to be in the job and how you are likely to perform.
Companies can now very easily build a picture of what makes, for example, the ideal salesman, specifically for their organization, based on success rates. This can then be used as a template for future hiring decisions.
Using Data Tails To Inform Decisions
An early experiment in this type of predictive recruitment was carried out by Xerox, which made surprising discoveries including the fact that previous experience was no indicator of success. Last year, their head of recruitment told the Financial Times: “I no longer look at somebody’s CV to determine if we will interview them or not.”
Cornerstone (formerly Evolv), which provides services to allow businesses to make data-powered recruitment decisions, routinely analyzes and tracks millions of appointments and employment actions. It then makes the data available (for a subscription) to allow businesses to build the profile of their perfect candidate, and match it against the applicants for the position.
The data is distributed anonymously but employers will match it against yours – in the hope that you will fit the pattern shown by others who performed similar roles successfully in the past.
The software has thrown up insights including the fact that long-term unemployed people often perform no worse than those who have been in continuous employment, and that the fact someone has “job hopped” (moved rapidly from one position to another) in the past is no indication that they will do so in the future. These may seem counterintuitive, but they are borne out by the facts.
These days of course, our social media profiles constitute a big part of the data trails we leave in our wake as we interact with the world, online and offline.
It has almost certainly been unwritten but common policy at many businesses to Google and Facebook search a potential new hire’s name to make sure there’s nothing lurking in the woodwork.
This can be good for picking up obvious problems, such as if the applicant is wanted for mass murder. But in normal circumstances anything that someone would want hidden is unlikely to appear on the front page of their Facebook feed.
But services are emerging which allow employers to assess a candidate’s suitability using more subtle indicators. As well as semantic analysis of the language used, patterns of behavior such as enthusiasm for subjects related to their fields can be monitored, and compared against patterns shown by previously successful employees.
Although Big Data is increasingly becoming a factor, the term “social recruiting” is still more frequently used to discuss the way businesses target their recruitment efforts – putting the right opportunities in front of the right people, rather than sifting candidates. However your data trail plays a big part here too, as keywords in your profile are used as indicators by algorithms which serve up the vacancy ads to the people most likely to apply.
Tinder for business?
CNN recently referred to solutions provided by the likes of Konetic and Hired as the “Tinder for Business”, in the way they are increasingly being used to matchmake between the best candidates and the most suitable positions.
Of course most employers will still rely on gut instinct to some degree – and I have always said that the best decisions are those that are informed by data and facts but interpreted through experience and common sense.
I also feel that the more senior the position you are applying for, the smaller the part that statistical modeling is likely to play in your selection. There are far fewer chief executive roles, so it’s difficult to build a significant enough data pool to compare candidates. However this is likely to change as companies which monitor employment markets extend the range of their data gathering.
There is always going to be a need for the human touch when it comes to recruiting. And it will be a long time, if ever, that we see the death of the job interview. But we are definitely going to see an increasing reliance on predictive modeling and social media analysis by recruiters. So if you’re looking for a job, you should take care to be in control of what your data trail says about you, as well as how neatly formatted your resume is.
This article is published in collaboration with LinkedIn. Publication does not imply endorsement of views by the World Economic Forum.
To keep up with the Agenda subscribe to our weekly newsletter.
Author: Bernard Marr is a globally recognized expert in strategy, performance management, analytics, KPIs and big data.
Image: Hays Recruitment Consultancy Section Manager Ignacio Ramos interviews Vicente Balmaseda at the Hays offices in downtown Madrid. REUTERS/Susana Vera.