Most Interviews Measure Polish, Not Performance
Long interview processes are exhausting. I’ve been on both sides of them, and I genuinely couldn’t tell you which is more stressful.
As companies try to hire more for skill and make better judgment calls on motivation, communication, and working style, the instinct is usually the same: add more steps, more panels, more hoops to jump through. It feels rigorous. It also does not seem to work very well.
Robert Half found 28% of new hires quit within their first 90 days. Usually not because they lacked technical skill, but because of judgment, motivation, communication, or working style. The problem is that most interview processes don't measure any of those things.
In-office hiring always had a crutch: ambient information. You noticed how someone carried themselves, handled small talk, or worked a room. None of that was especially reliable, but it felt like data.
Remote strips that away. Good. It forces a better question: what does this person actually produce when they work?
The answer is not another panel interview. It is a paid working interview.
PostHog pays candidates $1,000 for a one-day “SuperDay.” Doist uses paid test tasks pulled from real backlogs. Zapier has used paid trial projects. Different formats, same principle: pay people fairly to do real work, then evaluate the work itself.
That matters for candidates as much as employers. A paid trial says: we respect your time, we are serious about this process, and we are not asking you to perform for free. In a market full of ghosting and unpaid homework, that is not a small signal. It is the difference between a company that talks about trust and one that demonstrates it before the offer stage.
Compare that to the unpaid take-home that quietly became standard: four hours of work, a promise of careful review, then silence or a form rejection two weeks later. That isn't helpful for the process or respectful to the candidates.
It also selects for the people with the most spare time, not necessarily the best candidates. Parents, caregivers, and people already doing demanding jobs are more likely to opt out before you ever see what they can do.
Paid trials test what remote work actually requires: self-direction, written communication, async collaboration, prioritization, and the ability to ship under ambiguity. Those are the traits that determine whether someone succeeds at month six, not whether they aced a fourth Zoom panel.
The objections are predictable.
Yes, paid trials cost money. So does replacing a bad hire. SHRM estimates replacement cost at 50% to 200% of salary. The Work Institute pegged U.S. turnover costs at nearly $900 billion in 2023. The math is not close.
No, top candidates are not scared off by fair evaluation. They are scared off by unpaid speculation and messy processes.
Start small. Pick one role you have struggled to hire well. Scope a project that mirrors the actual work. Let candidates complete it on their own schedule. Pay fairly. Build the rubric before the trial starts. Close the loop quickly.
The traditional interview was designed for a world that overvalued chemistry and proximity. Remote work exposes that weakness.
That is not a hiring problem. It is a chance to build something better.
