I could tell right away from the tone of his voice that the VP of Engineering wasn’t happy. He practically growled at me. He had just finished interviewing a job candidate named Anand, who I had directed his way, and was calling me to say he was going to pass.

Just a few minutes earlier, Anand had called and raved about how well the interview had gone. He had interviewed for nearly a full day, meeting with different leaders across the organization, including, at the end of the day, the VP of Engineering. I had helped this company build out a new “platform” strategy, which is why I was trying to identify the right candidates to work on it, and I thought Anand would be a great fit.

But the VP and Anand had strikingly different reports about their meeting. Anand said that he had asked far more questions than he usually did, asking for detailed and specific information on the strategy that helped him understand the complexity of the challenge the company was facing. He felt like he had engaging, insightful conversations with everyone he met. In contrast, the VP told me that he found Anand’s questions “super annoying.”

This isn’t the first time I’ve heard a leader say that a perfectly qualified candidate is a “bad fit.” Candidates are too often screened out because they don’t fit a particular pattern – one survey found as many as 75% of resumes don’t make it past Applicant Tracking Systems. As I discussed the issue further with the VP, I learned that he thought that Anand had the right skills and experience but that he found Anand’s questions annoying. He said: “He asked us a ton of questions that the team didn’t have the answers to.” His assessment that Anand was a “bad fit” was really code for “I don’t want to feel uncomfortable.”

Innovation requires not knowing long enough to learn new things. How can you build something new, if you aren’t okay with not already knowing the answer? The future is not created; it’s co-created. Leaders need to build teams that can both define the right questions, and then discover new answers.

Instead of being annoyed by Anand’s questions, the VP should’ve welcomed them — and asked Anand questions in return. That is, of course, the value of an interview. An employer seeks to learn about the candidate’s skills and relevant experiences. And a good candidate uses questions to learn about the role, the boss, and the company to assess whether it’s the right job. Here are some types of questions the VP might’ve asked — and the ones you should ask — to avoid screening out a perfectly good candidate based on the wrong criteria.

Questions that uncover capabilities, not just experience. Are you asking questions that get to someone’s capabilities or are you seeking confirming data that someone has done exactly what you have already scoped? Several years ago, a colleague asked if I’d review their job description for a social media “expert.” Twitter had been around for maybe a year at this point and when I looked at the description, I just started laughing. The first line read “10 years of experience.” Quite often, we use useless metrics to scope a job to do what has already been done. The upside of asking for years of experience is we get someone who has done what we need. The downside is we risk limiting what we can create next by doing what has already worked. Instead of asking, “Have you done x or y or z?” you want to ask, “How would you approach doing x or y or z?” This shift in question lets you learn someone’s capacity to think with you.

Unfortunately, right now, an estimated 77% of all jobs (60% in the U.S. and 80% worldwide) require little to no creativity, decision-making, or independent judgment. But if you are working on innovation, you need someone who can think with you. And by focusing on capability over experience, you increase the chances you find that person.