Empowered by job-seeking sites and online career boards, candidates no longer have to diligently type out and mail resumes and cover letters to the jobs they’re genuinely interested in. Instead, with a single click, they can apply to numerous jobs at a time. For job seekers, this makes things easier and saves enormous amounts of time, but for businesses, it’s created an overabundance of documents — resumes and cover letters in quantities that would have been unthinkable a few decades ago.

The result is that HR departments are increasingly reliant on applicant tracking systems (ATS) to process this influx of applications. Of course, these systems do have their advantages, and given the state of the internet, ATS technology seems here to stay. But as companies farm more and more of their recruiting and resume screening out to automated systems, they’re discovering the hard way that these systems aren’t infallible — and that a human touch is still necessary in the hiring process.

Despite their convenience, increased use of ATS technology may be inadvertently fast-tracking less qualified job applicants, causing companies to miss out on some of the best candidates. Because they operate based on keyword searches and other measurable metrics, the overall quality of the candidates that ATS promotes is hardly guaranteed.

Often, the first barrier a candidate has to cross in order to be considered seriously for a position is not an HR professional, but a series of program inputs that may or may not place the best applicants at the top of the list. ATS technology parses resumes and cover letters and searches for keywords that automatically disqualify applicants, discarding their resumes without a pair of human eyes ever seeing them. These systems can throw out hundreds of resumes from potentially qualified applicants simply because they contained (or lacked) a single keyword!

It’s possible, for instance, that a preprogrammed ATS may jettison any resume that doesn’t include a master’s degree. The unintended consequence of this is that any candidate with comparable (or even more relevant) prior experience, master’s degree or no, will be rejected. Furthermore, not every applicant with a master’s degree will be qualified for the role — yet they may be pushed through anyway.

Accordingly, every stack of applications that makes it to a hiring manager’s desk will include people who are unqualified or a poor fit for the role. In computer science, this is referred to as the “garbage in, garbage out” problem: if the instructions given to an ATS are flawed, and if the vast majority of resumes received come from unqualified candidates, there’s a huge chance that the applications that make it through the initial screening process will be just as flawed as the original batch of resumes received.

The irony is that rather than streamlining the hiring process, over-reliance on ATS technology unnecessarily protracts it, costing companies time and money. Hiring relies heavily on intangibles like “personality” and “fit” for a reason: humans, unlike an automated system or an algorithm, understand candidates’ personalities as well as company culture. They can determine how well a candidate will fit in with a team, as well as what specific value they can add to a business.

Online job postings and career boards aren’t going away anytime soon, so for better or for worse, ATS will remain a necessary component of any company’s hiring process. However, it’s important to remain skeptical of these systems; don’t blindly entrust too much power to them, or cut out the human element in the process entirely. It still takes a human being to understand another human being — which is why people are more important than ever for companies who want the right human being for the job.