Hiring in tech is challenging.
In a ‘high demand” market, recruiters are struggling to identify and recruit talent.
But, even when they do, studies have shown that within a year of being hired, an astounding 84 percent of new hires will not live up to expectations.
This means that in most cases these new hires end up being mistakes that can cause more harm than good. According to Dr. Bradford Smart’s estimate, the average mis-hire earning $100,000 will end up costing $1.5 million. For some industries, like the tech industry, such hiring mistake can cost even more.
In this two-part article we will examine some of the common mistakes made by human recruiters and how they can be solved with AI.
Your perfect candidate is not actively “looking for a job”
As any headhunter will tell you, in many cases the “top-talent” candidates are often not actively looking for a job. This means that simply posting a job will not get their attention.
According to Liran Kotzer, Woo’s CEO “You don’t want to search for a job when you need to search for a job. You want to search when there is a better opportunity for you.”
Head hunters try to reach these talents and discreetly offer them “the right opportunity”. But in the age of AI-based recruiting, like the one offered by Woo.io, candidates can anonymously tell the AI their “wish list” that will make them move from their “comfort zone” and then the AI finds a matching opportunity.
“This wish list is not only about monetary compensation…” said Kotzer, “it can include a wide variety of parameters, including location, company type, work-life balance, etc. Candidates are willing to share their aspirations because they can remain anonymous throughout the entire process.
And companies that are willing to match these aspirations can do so without the discomfort associated with the head hunting process.
Job description about “skills” and experiences”
In most cases job descriptions are based on a list of skills and years of experience. Of course, this is a good way to screen out completely unqualified candidates, as someone with this set of skills and experience is more likely to be a good fit.
But, these traditional competence qualification descriptions miss non-traditional candidates who can do the work, but have a different set of skills and experiences.
These questions don’t necessarily answer the most important question.
“will they be able to perform well on the job?”.
To answer this question, instead of focusing on competence qualifications, recruiters should also include performance-based job description or questions.
For example, it is possible to include a set of performance objectives, such as “Completed the PRD for a logistics system in less than 30 days”. Not only is this more engaging and informative, it’s also more accurate.
Performance-based descriptions get to the heart of the job and open the door to more diverse applicants. The problem is that this type of recruitment is much more complicated and requires processing of much more data.
If the competence qualification was a based on “Boolean” questions, such as “Do you have at least 5 years’ experience in PHP?” (“yes” or “no”), the performance data presents a more complex picture.
For example, what if the candidate did not complete the PRD in 30 days, but has completed the entire project before the deadline? or has received a high NPR (Net Promoter Score) for the app that he/she developed?
A highly skilled recruiter can analyze the performance-related answers of a single candidate (ok, even of 10), but they cannot do it for 10,000 applicants, right? This is where AI comes in.
The AI can represent the thinking or the logic of the human recruiter and amplify it so it can process 10,000 or even 100,000 applicants.
The AI can analyze and learn from massive amounts of data. So, for example, when AI wants to propose matches to some candidate, it can go over all the candidates in the database and find candidates similar to that candidate that was selected by the human recruiter.
In the next part we will continue our examination of the common mistakes made by human recruiters that can be solved with AI.
