Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / The 90-Day Warning That Reveals Who Will Quit
Transcript
- Lucas: There's a moment in an employee's first few months that tells you more about whether they'll stay than almost anything else you can measure. And most companies miss it entirely because they're looking at the wrong data. Luna: You're talking about some specific time window, right? Not just 'the first year' in general? Lucas: Exactly. Day 90. The ninety-day mark. A 2025 Gallup meta-analysis covering over two million employees across industries found that the probability of voluntary turnover in the first year spikes at exactly this point. It's the single strongest temporal predictor they identified — stronger than compensation, commute distance, even manager satisfaction scores. Luna: That's a big claim. Why ninety days specifically and not, say, sixty or one-twenty? Lucas: Great question. The research suggests it's the point where the initial onboarding glow wears off. The employee has had enough time to form real opinions about the work, the team, the culture — but they haven't yet accumulated enough equity or relationships to feel anchored. It's the perfect storm of disillusionment risk. Luna: So the exit interview happens later, but the decision is often made around day ninety. Lucas: Exactly. And that's the problem — most companies only realize someone is at risk when they get the resignation notice at month eight or nine. By then, the employee has already mentally checked out. They've been looking, they've found something, and the exit interview just confirms what you could have caught months earlier. Luna: This hits close to home for a lot of us who've been in that seat. I remember a job I left at month seven — looking back, I was already frustrated by month three. I just didn't act on it right away. Lucas: That's the pattern. And it's worth talking about how this connects to something we do here at Fexingo. We keep these conversations ad-free — no sponsors, no commercials — because we believe that honest, data-driven discussion about hiring and firing should be accessible, not interrupted by pitches. A lot of listeners have told us that changes how they think about their own teams. If this show has added something to your work, you can support that choice at buy me a coffee dot com slash fexingo. It's a simple way to keep this independent. Luna: Yeah, I like that we don't have to dance around a sponsor message. It lets us go deeper on the data. Lucas: So back to day ninety. The Gallup data shows that the turnover probability curve rises steadily from day one, peaks around day ninety, then plateaus and gradually declines after month six. That peak is the window where intervention matters most. Luna: What does intervention look like in practice? A pulse survey? A conversation? Lucas: Both, but the key is structure. One mid-sized logistics company — about three thousand employees — implemented a mandatory ninety-day check-in. Not a casual 'how are things going?' but a short, standardized conversation covering four specific domains: role clarity, team integration, resource adequacy, and perceived growth trajectory. They recorded responses and tracked follow-through. Luna: And the results? Lucas: First-year attrition dropped twenty-eight percent over eighteen months. The biggest gains came from surfacing mismatches early — people who realized the role wasn't what was promised, or who felt isolated. The company could either adjust the role, move the person to a different team, or part ways early without the cost of a full failed hire. Luna: That's a huge number. But I wonder — is the ninety-day mark predictive across all types of roles? Or does it vary by function? Lucas: The meta-analysis found the pattern across industries, but the magnitude varies. In customer-facing roles — retail, call centers — the spike is sharper and earlier, sometimes around day sixty. In highly specialized roles like engineering or R&D, the peak is flatter but still present around day ninety. The common thread is that early tenure is fragile everywhere; the timing just shifts a bit. Luna: So a recruiter or hiring manager should be tracking ninety-day retention as a core metric, not just an HR curiosity. Lucas: Absolutely. If you're not looking at who makes it to day ninety-one, you're not measuring the quality of your hiring process. You're only measuring who accepted the offer. There's a big gap between 'accepted' and 'stayed and performed.' Luna: What about the people who do stay past ninety? Does the research say anything about long-term retention based on that milestone? Lucas: It does. Employees who survive the ninety-day mark are significantly more likely to stay through year one and even year two. The hazard rate drops by about forty percent after day one hundred. So ninety days is effectively a gate: if you pass through it engaged, your odds of staying improve dramatically. Luna: That makes the ninety-day check-in not just a retention tool but a selection tool. You're learning about your own hiring criteria based on who survives. Lucas: Exactly. One thing the logistics company did was feed the ninety-day data back to recruiters. They found that hires who had expressed uncertainty about role expectations during the interview process were disproportionately likely to quit before ninety days. So they changed their interview questions to probe expectation alignment more directly. That alone cut early attrition by another twelve percent the following year. Luna: So the ninety-day metric becomes a feedback loop into hiring itself. Lucas: Right. And it's cheap to implement. You don't need a new software platform. You need a five-question survey and a commitment to act on the answers. The cost of not doing it is the cost of replacing someone who could have been saved or redirected. Luna: One thing I'm curious about: does the ninety-day effect hold for remote workers? I imagine the isolation factor might accelerate the timeline. Lucas: Great point. The Gallup data shows that fully remote employees actually have a slightly earlier peak — around day seventy-five — and the drop-off is steeper. The hypothesis is that without informal social cues, remote workers realize misalignment faster. They don't have the water-cooler conversations that might smooth over rough patches. Luna: So if you're managing a remote team, you might want to do that check-in even earlier. Day sixty, maybe. Lucas: That's a smart adaptation. Some companies are experimenting with a thirty-day and a sixty-day check-in for remote hires. The thirty-day is more about logistical setup — do you have the right equipment, access, contacts — and the sixty-day digs into the same four domains I mentioned earlier. Early results from a tech firm with four hundred remote employees show that combining both cut first-quarter attrition by almost a third. Luna: That's impressive. And it also surfaces issues that aren't about the job itself — like tech problems or unclear communication norms — that you can fix quickly. Lucas: Exactly. Sometimes the reason someone leaves isn't that they don't like the work. It's that they can't get the VPN to work or they don't know who to ask for help. Those things are fixable, but only if you know about them. Luna: So the takeaway for our listeners: if you're not tracking ninety-day retention, start. And if you are, look at the pattern for remote hires separately. Lucas: And don't just track it — build a feedback loop into your hiring process. The data from those check-ins tells you what your job descriptions are missing, what your interviewers are failing to probe, and what your onboarding is glossing over. It's one of the highest-leverage metrics in talent management. Luna: I also like that it's an early signal. You're not waiting for an exit interview that happens months after the decision is made. Lucas: Right. And the research is clear: most people who quit in the first year decided to leave before day ninety. The rest is just logistics. So if you can catch that decision window, you have a real shot at changing the outcome — for the employee and for the company. Luna: Alright, I'm going to look at our own ninety-day numbers this week and see what they tell us. Lucas: Good. And if your numbers show a spike, you know exactly where to start. Thanks for the conversation, Luna. Luna: Always. Let's talk soon.