This blog post is based on the second episode of the data-driven recruiting podcast hosted by CodeSignal co-founders Sophia Baik and Tigran Sloyan. You can find and listen to this specific episode here or check out the video version embedded below. If you ask companies what their biggest challenge is today (when it comes to recruiting anyways), they’ll tell you it’s the top of the funnel. AKA, not enough “good” candidates are applying to their jobs. But the truth is, more and more people are entering the tech industry and job market. So, who’s right – the companies saying they don’t have enough candidates or the data saying there are more applicants than ever? Turns out, it’s both. But it doesn’t come down exclusively to numbers. It comes down to recruiting. Normally more than 90% of candidates are eliminated at the resume review stage – and that review usually consists of basic keyword searches. That may have worked 30 years ago, but today – rejecting 90% of your applicants with such rigid limitations is like putting handcuffs on before entering a ring. It’s probably not going to end well. Today, qualified candidates are learning their skills online – more than 50% of developers are self-taught – and they probably don’t have a degree from those schools you were searching for. It’s not that they couldn’t get into them. It could be, they didn’t want to. Universities are lagging when it comes to the tech field, and candidates are finding their own way to get what they need to be successful. Plainly speaking, education is how one acquires skills and recruiting is how you look for those skills. To work properly, these two processes have to evolve together. But as high-quality education advances away from top tier universities, recruiting has to adjust. Traditional education routes and pedigrees are old proxies that need to be retired. Some people have recognized this and tried to identify new proxies. Like testing a candidate’s skill level, for example. But that data isn’t exactly perfect. And it too can become quickly outdated. So, what works? Testing a hard skill tends to be direct and accurate. A hard skill can be understood as a direct measure of ability. Can you pick up this weight today, for example? If so, we believe you will be able to pick it up after being hired. But is it enough? Skeptics ask if it properly reflects ambition and drive, like getting accepted to a top school may have indicated in the past. Google asked the same question when evaluating its workforce. In the early days, they decided to expand their hiring to include candidates from second and third level schools – not just the Harvards and Stanfords. Working with professors to identify the best candidates from each class, they branched out. They made hires and then tracked them against those with more traditionally-prized pedigrees. If you guessed that the new hires did better in the long run, you’d be right. A clear indicator that it’s time for a change (You can learn more about this story from this video). So, what’s the bottom line? To grow the list of qualified applicants, start with measuring talent at the top of the funnel, rather than eliminating 90% of them with outdated practices. Measure skill directly and you may be surprised how many great candidates you’ll have to consider for your next job opening.