Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / What a Three-Month Probation Actually Reveals About New Hires
Transcript
- Lucas: So there's this logistics firm in Chicago, about 400 employees, mostly warehouse and dispatch roles. They hire around 90 new people a year. And for years, their probation period was basically a formality. Luna: Right — three months of just letting the person settle in, hoping it works out. Lucas: Exactly. And they were losing about a third of new hires in the first year. Not because they were bad recruiters — but because they had no real system for evaluating fit during that probation window. Luna: So they redesigned it. I've seen this approach before — making probation an active diagnostic rather than a passive waiting period. Lucas: They called it their '90-day onboarding diagnostic'. And the core idea is simple: every week, the new hire and their manager have a fifteen-minute structured check-in. Not a status update — a set of five questions, the same questions every week. Luna: What kind of questions? Lucas: Things like 'What part of the role felt most natural this week?' and 'Where did you need help that you didn't get?' And the manager answers a parallel set — like 'On a scale of one to ten, how confident are you that this person will succeed in this role?' They track the scores over time. Luna: That's interesting — the manager's confidence score as a metric. Because people are often slow to admit a hire isn't working. Lucas: Right. And that's the whole point. They found that by week eight, the manager's score was a remarkably reliable predictor of whether the person would stay past year one. If the score dropped below a 6 by week eight, there was an 80 percent chance the person would leave or be let go within six months. Luna: So the third month becomes the real tell. Weeks nine through twelve, you're basically confirming the pattern. Lucas: Precisely. And they built something else into this: skill-based milestones. By the end of month one, you need to demonstrate you can handle the core operational tool — in their case, a warehouse management system. By month two, you need to resolve a customer issue without escalation. By month three, you train someone else. Luna: That's really smart — the training milestone. It forces the new hire to internalize the knowledge enough to teach it. Lucas: And it reveals a lot. Some people are great at doing the work but can't articulate it. That doesn't mean they're a bad hire — but it tells you where they'll struggle if you want them to grow into a team lead role. Luna: What did the data show after they implemented this? Lucas: First-year attrition dropped by 34 percent. But more interestingly, the average time to flag a problem hire went from about 14 weeks down to 7. So they were intervening — or deciding to part ways — much earlier, which saved them months of salary and training cost. Luna: There's also the manager side. I've seen studies showing that managers often wait too long to act on a bad hire because they don't trust their own judgment. Having a structured framework gives them that confidence. Lucas: Yeah, that's a huge piece. The company reported that after the change, managers felt more confident in their hiring decisions overall — not just during probation, but in who they chose to hire in the first place. Because they knew they'd have a real system to catch mistakes. Luna: So the probation period becomes a feedback loop that improves hiring itself. Lucas: Exactly. The HR team started looking back at the check-in data to see which interview questions correlated with high week-eight scores. They found that candidates who gave specific, concrete examples of adapting to a new process in a previous job — not just 'I'm adaptable' but 'In my last role, I learned a new inventory system in two weeks' — those people scored significantly higher. Luna: So they changed their interview questions based on probation data. That's a closed loop. Lucas: It is. And it's rare. Most companies treat probation as a one-way evaluation of the employee, not as data that can improve the hiring system. Luna: What about the employee's perspective? Did they survey new hires? Lucas: They did. And interestingly, new hires reported feeling more supported, not more surveilled. Because the weekly check-in was framed as a coaching conversation, not a test. The manager was saying 'I'm here to help you succeed, and I need to know where you're struggling so I can help.' Luna: That's a crucial framing. If you call it a probation 'review', it feels punitive. If you call it an onboarding 'diagnostic', it feels like support. Lucas: Right. And the company actually stopped using the word 'probation' internally. They called it the 'onboarding partnership period'. Language matters. Luna: So what's the one thing a smaller company — say 50 people — should steal from this? Lucas: The weekly check-in template. It's five questions, takes ten minutes. You don't need a full HR system. Just a shared doc where the manager and new hire both fill it out each week. The key is consistency — doing it every week, not skipping weeks when things are busy. Luna: And the manager's confidence score question. That's the hidden gem. Lucas: Absolutely. Because it forces the manager to be honest with themselves. If by week six you're still at a 7 but you can't articulate why not higher, that's a signal. Luna: If today's conversation gave you a practical idea to try with your next hire, that's exactly what we hope for. And if you find that these conversations are moving your work forward in some small way, listener support is what keeps this show ad-free and independent. You can find it at buy me a coffee dot com slash fexingo. Lucas: Yeah, it's a simple way to keep the episodes coming without any sponsors or interruptions. Really appreciate the people who do that. Luna: Alright, back to the diagnostic. One thing I want to highlight — this company also built a 'go/no-go' decision point at the end of month two. Not at month three. Lucas: That's smart. Because by week eight, they had enough data. Waiting another month just delayed the inevitable. They would sit down with the manager and say 'Based on the weekly scores and milestone completion, is this a go or no-go?' If it's a no-go, they'd start a performance improvement plan or part ways by week ten. Luna: So the actual probation period effectively became two months, not three. The third month was just for confirming and solidifying. Lucas: Exactly. And that saved them money — they weren't paying someone for an extra month when they already knew it wasn't working. Luna: There's a behavioral economics angle too — the sunk cost fallacy. Managers who wait three months feel like they've invested so much that they're reluctant to let someone go. Having a hard decision point at week eight kills that trap. Lucas: Right. And the company made that decision point non-negotiable. Even if the manager was on the fence, if the scores and milestones said no-go, they had to escalate to HR. It took the emotion out of it. Luna: Let's talk about the milestones themselves. You mentioned three: core tool proficiency, customer issue resolution, and training someone else. Those are pretty specific to logistics. What would that look like for, say, a marketing agency? Lucas: You'd map it to the critical skills for that role. For a content writer, month one might be 'publish two blog posts with minimal edits', month two 'conduct an interview and produce a piece from it', month three 'mentor a junior writer on one piece'. The principle is the same — it's about demonstrating not just competence, but the ability to learn and teach. Luna: And the milestones should be observable. Not 'understand our brand voice' — that's vague. But 'write a piece that requires no tone edits' is observable. Lucas: Exactly. The company learned that lesson the hard way. Their first version of milestones were too fuzzy — 'demonstrate understanding of our safety protocols' — and nobody could agree on whether someone had passed. They had to rewrite them to be binary: either the person did the thing or they didn't. Luna: One more thing — how did they handle the case where someone passed all milestones but the manager's confidence score was still low? Lucas: That happened in about 15 percent of cases. And usually it came down to cultural fit or communication style. The person could do the job but didn't mesh with the team. The company's policy was that if the milestones were met but the score was below 6, they'd try a team rotation — move the person to a different department for a trial period. Luna: That's a compassionate approach. Not just 'you're fired' but 'maybe you belong somewhere else here.' Lucas: And it worked. About a third of those rotated employees ended up staying long-term in their new role. So the diagnostic wasn't just about cutting losses — it was about finding the right fit within the company. Luna: That seems like a better outcome than losing someone entirely. And it's cheaper than rehiring. Lucas: Much cheaper. And it builds a culture of internal mobility. Employees see that the company is willing to invest in finding the right role for them. Luna: Alright, so the takeaway for anyone listening: if your probation period is just a waiting game, you're leaving money on the table. Steal that weekly check-in template, add skill milestones, and set a hard decision point at week eight. Lucas: And if you're a manager, start by asking yourself one question each week: 'On a scale of one to ten, how confident am I that this new hire will succeed here?' Track it. You might be surprised what you learn.