The Math On AI For Recruiting: It Works!

An Excerpt of The Rise of Talent Intelligence by Talent Tech Labs Download Banner

Have you been curious as to if the numbers and costs associated with AI for recruiting would be positive in your company?

The discussions about AI in the recruiting process have been robust in 2018 and 2019. More and more companies are adopting these solutions, or some form of machine learning and artificial intelligence, but there’s a lot of companies out there still doing things in an old-school, manual way. That makes sense, obviously — change is not easy, especially as an organization gets bigger and there are more stakeholders to inform and get approval from — but the math is not on the side of old-school recruiting approaches staying as the norm for long. 

Think of the numbers

An average large-size company job posting gets about 250 applications. Delta had 200,000 applicants for one flight attendant position this year. Google gets about 2 million applications per year. But it’s not just big places. Netflix, for example, only has about 3,000 employees — but receives well north of 100 applications for almost every job. Strong brand name companies, almost regardless of size, are viewed as good places to work on the open market. The resumes roll in.

Now bring in the idea that recruiters scan resumes for about six seconds. That’s the conventional number we’ve used for much of the last 10 years, and even if you think it’s a little higher — let’s say they are scanning a LinkedIn profile and that might be 10 seconds or so of attention — still the number of seconds a recruiter spends on an initial scan of a new candidate is very small.

Ideal Blog Banner

For ease of math, let’s say 10 seconds per candidate. With 250 applicants at a large company, you’re now at 2,500 seconds of recruiter time. That’s about 42 minutes, if you round up. That’s 8% of the workday gone just on screening. 

On face, this might not feel like a lot — but remember, these numbers are for one open position. Companies like Amazon periodically have upwards of 30,000 jobs open.

You probably won’t have 30,000 jobs open right now — most companies are not Amazon — but even if you have 10 or 20, it becomes increasingly impossible to scale the top-of-funnel recruitment activities without using technology in some way. 

Screening is the first brick in the building of successful recruiting

If your screening is rushed or flawed, it’s hard to fix that in later stages of recruiting — because now the wrong people are getting in front of hiring managers for interviews, and if the hiring managers don’t like them, the whole process will go back to the top of funnel. That’s costly and not a good use of time, especially if the roles are urgent for a current project. 

Rather than taking up 8% or more of a recruiter’s day, hand that side of the process over to technology. Artificial intelligence has the potential to give you a 100% screen rate (!) — whereby every candidate is screened — and all those candidates will be sorted as A, B, C, or D level. It’s an easy way to organize information about prospective candidates and sort it for other stakeholders, i.e. the hiring manager. Plus, your feedback about previous quality of hire informs the software going forward. This allows it to essentially “get smarter” and find the right fit for your organization, which helps with retention down the road. It’s important for the tech side of the recruitment process not to end with hiring itself, because quick turnover is a big cost to businesses too. What you’re looking for are good fits that can contribute to key work quickly, but people who will stay for years as well. Then you’re not constantly trying to fill open roles.

That’s what our software is designed to do: Save you time, save you money, find you great people, and find you people who will stay. 

And from a sheer numbers perspective, doesn’t that make logical sense? Take that 25%+ of your workday and put it back towards having actual conversations with candidates and hiring managers to get a better sense of role, future needs, and potentially great fits (on the candidate side) where there’s not necessarily the right open role yet. Be a conversationalist and idea-generator, not just a screener. Tech can do that better right now anyway!

Let us know if you have any questions.