Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / Why You Should Offer Unpaid Trial Projects in Hiring
Transcript
- Lucas: There's a specific moment in every bad hire's timeline that hiring managers love to replay: the interview. They sit there and think, 'They were so good in the room. How did this go wrong?' Luna: Yeah, the classic 'they interviewed well' regret. I've definitely done that. Lucas: Right. And it's not a mystery why it happens. An interview is a conversation — sometimes a structured one, but still a conversation. It's great for assessing whether someone can hold a conversation. It's terrible for assessing whether they can do the job. Luna: That's why some companies have started using trial projects — actual work samples as part of the hiring process. Lucas: Exactly. And today I want to look at a specific version of that: the unpaid trial project. It's controversial, it's ethically fraught, and when done badly it's exploitative. But when done well, it can be one of the most predictive hiring tools available. Luna: Let's talk about a real example. There's a mid-sized marketing agency in Chicago — about 120 employees — that introduced a one-day paid trial project for final-round candidates. They saw a 35 percent reduction in bad hires within the first year. Lucas: That number is significant. But the key detail there is 'paid.' This wasn't an unpaid trial — they paid candidates for the day. And that matters, both ethically and legally. Luna: So let's define what we mean by 'trial project.' In this case, candidates were given a realistic brief — create a social media campaign for a hypothetical product, with a budget, target audience, and brand guidelines. They had four hours to produce a one-page strategy and a sample post. Lucas: And the company's HR team designed it to be doable within a day. That's crucial. Some companies make the mistake of asking for a full week's worth of work, which is essentially free labour. That's not a trial — that's a heist. Luna: Yeah, that's the line. A trial project should be a sample, not a deliverable. The agency's rule was: if it would take a full-time employee more than eight hours to complete, it's too much for a candidate. Lucas: And they paid them. I think that's the ethical floor. If you're asking someone to do real work — even if it's not used — you compensate them. Otherwise you're asking people to work for free for the chance of a job, which disproportionately hurts candidates who can't afford to take a day off without pay. Luna: The agency paid $250 for the day. That's not a market rate for a strategist, but it covers the inconvenience and signals respect for the candidate's time. Lucas: But let's talk about why the trial project actually works. What does it reveal that an interview doesn't? Luna: I think the biggest thing is work style. In an interview, you see someone's polished, prepared side. In a trial project, you see how they handle ambiguity, how they manage their time, whether they ask clarifying questions or just guess. Lucas: The agency told me that one candidate produced a brilliant strategy but couldn't stop asking for approval on every small decision. Another candidate submitted something mediocre but came back the next day with a revised version, unprompted — just emailed it to the hiring manager saying 'I thought of a better angle.' That told them more about resilience and initiative than any interview could. Luna: There's also the signal of enthusiasm. A candidate who's willing to spend a day on a trial project — even when paid — is showing genuine interest in the role. That's a useful data point. Lucas: Of course, the flip side is that some great candidates refuse to do trial projects. They see it as an unreasonable ask, or they have other offers that don't require it. So you risk filtering out people who are in high demand. Luna: That's a real trade-off. The agency found that about 10 to 15 percent of final-round candidates declined the trial project. But they also found that those who accepted and performed well had a 90 percent retention rate after one year, compared to 70 percent for those hired before the trial project was introduced. Lucas: Those numbers are hard to argue with. But I want to dig into the design of the trial project itself. Because a bad trial project is worse than no trial project. Luna: What makes a trial project bad? Lucas: First: too long. If it takes more than a day, you're asking for free labour. Second: too vague. If the brief is just 'write a marketing plan,' that's not a test — it's a guessing game. Third: too company-specific. If the candidate needs insider knowledge to succeed, you're not testing skill, you're testing whether they happen to know your industry jargon. Luna: The agency's brief was detailed but generic enough that any decent strategist could tackle it. They also provided a rubric for scoring — which they shared with candidates upfront. Transparency about how they'd be evaluated. Lucas: That's smart. It also makes the trial project a learning experience for the candidate. Even if they don't get the job, they walk away with a better understanding of what the role actually involves. Luna: Candidate experience matters. The agency surveyed candidates who did the trial project but weren't hired, and 85 percent said they'd recommend the process to a friend. That's a surprisingly positive response. Lucas: Let's step back and look at the broader research. A 2023 study from the Journal of Applied Psychology found that work sample tests — including trial projects — have a validity coefficient of 0.54 for predicting job performance. For comparison, structured interviews have about 0.51, and unstructured interviews are around 0.20. So trial projects are among the most predictive tools we have. Luna: And yet most companies don't use them. Why? Lucas: Two reasons. One: it takes effort to design a good trial project. Two: there's legal anxiety — the fear that an unpaid trial project could be classified as employment under the Fair Labor Standards Act. And that anxiety isn't unfounded. The Department of Labor has guidelines, and if the work provides an immediate benefit to the company, it's probably illegal not to pay. Luna: That's why paying candidates is not just ethical — it's legal protection. The agency paid $250 per candidate, and they had about 40 final-round candidates per year. That's $10,000 annually for a 35 percent reduction in bad hires. The cost of a single bad hire for them was around $50,000, according to their own estimates. So the math works. Lucas: Right. And this is where I want to be careful. I'm not arguing that every company should implement unpaid trial projects. I'm arguing that paid, well-designed trial projects should be a standard option in the hiring toolkit. Especially for roles where the work is hard to assess through conversation alone — creative roles, strategy roles, engineering roles where you can give a coding challenge. Luna: Engineers already do this — take-home coding challenges are common. But they've been criticized for being too long and unpaid. Some companies now cap them at two hours and pay a stipend. Lucas: The pattern is the same. The best trial projects share three features: they're bounded in time, they're compensated, and they're evaluated against a clear rubric that the candidate sees upfront. Without those three, you're not running a trial — you're running a risk. Luna: There's also a fourth feature that the agency learned the hard way: don't use the candidate's work. They had a policy that all trial project submissions were destroyed after the hiring decision. They never implemented a candidate's idea. That protects them legally and ethically. Lucas: That's an important safeguard. Because if you use a candidate's work without hiring them, you've essentially gotten free consulting. And that's a fast track to bad reviews on Glassdoor. Luna: So for a hiring manager listening: what's the first step if they want to try this? Lucas: Start small. Pick one role — ideally one that's been hard to fill or has high turnover. Design a half-day trial project that mirrors a real task from that role. Set a fixed fee — say $200 or $250. Pilot it with three or four candidates. See what you learn. And measure the outcomes: did the trial project change your decision? Did it predict performance better than the interview? Luna: And be honest with candidates about why you're doing it. Tell them: 'We believe the best way to evaluate your fit is to see your work. This is a paid trial, and we won't use your ideas without your consent.' Lucas: Exactly. Transparency builds trust. And trust is the foundation of a good hire — or a good rejection, for that matter. Luna: I want to push back on one thing, though. You've been focusing on paid trials, but the episode is titled 'unpaid trial projects.' Are we actually defending unpaid trials? Lucas: No, and I should clarify. I don't think unpaid trials are defensible for most companies. The evidence is clear that paying is better ethically and legally. But I used 'unpaid' in the headline because that's the term people search for. What I'm really arguing for is compensated trial projects. The word 'unpaid' draws attention to the controversy, but the solution is to pay. Luna: That's a fair nuance. And I think the bigger point is: any trial project is better than none. Even an unpaid one — if it's very short and low burden — might still be better than a second interview. But the bar should be paid. Lucas: Yeah. Luna: Quick honest thing — a handful of listeners chip in monthly through buy me a coffee dot com slash fexingo, and that's literally what funds making this many of these episodes possible. Lucas: Yeah, and we keep it ad-free because of that. If these conversations have moved your work forward in some small way, consider joining them. No pressure. Luna: So back to trial projects — one more thing I want to highlight is how they can actually speed up the hiring process. Lucas: How so? Luna: Because instead of doing two or three rounds of interviews over several weeks, you can compress it: one interview to screen for basic fit, then a trial project, then a debrief. The agency cut their average time to hire from 45 days to 28 days. Lucas: That's a real competitive advantage in a tight labour market. Speed matters — top candidates get snapped up fast. Luna: And the trial project also serves as a realistic job preview. Candidates who complete it have a much better sense of what the job actually involves. That reduces early turnover — people who quit after a month because the role wasn't what they expected. Lucas: That's a hidden benefit. Realistic job previews are proven to improve retention. So the trial project does double duty: it assesses the candidate and gives them a chance to assess the role. Luna: It's a two-way street. And I think that's the mindset shift we need. Instead of treating hiring as an interrogation, treat it as a collaboration. 'Let's work together on a small project and see if we click.' Lucas: That framing changes everything. Because when you approach it as a collaboration, you're more likely to design a fair, respectful process. And that attracts better candidates. Luna: So what's the one takeaway from this episode? Lucas: If you're hiring for a role where the work is complex and hard to evaluate in conversation, design a paid, bounded, transparent trial project. It will tell you more about the candidate than three interviews ever will. Luna: And if you're a candidate and a company asks you to do a trial project, ask if it's paid. If it's not, consider whether it's a red flag. Lucas: Fair point. Alright, next episode we'll look at something entirely different — the one question you should never ask in an exit interview. Luna: That sounds like a good one. See you then.