The Boss Who Knows You're Quitting Before You Do — predictive…

Your manager pulled you aside last Tuesday. Nothing was wrong, she said. She just wanted to check in. But here's the thing—she already knew. Before you updated your LinkedIn. Before you took that "doctor's appointment" that was really a second-round interview. Before you even admitted it to yourself. She knew because a dashboard told her. Welcome to the strange new world of predictive intelligence targeting teams—software that scores employees on how likely they are to quit, slack off, or blow up a project, often weeks before any human notices a thing. Companies like Workday, Visier, and a growing crop of startups now sell "workforce analytics" that crunch your emails, your meeting attendance, your badge swipes, and your performance reviews into a single number. Your flight risk. Your engagement score. Your predicted value to the team over the next 90 days. It sounds like science fiction. It's a Tuesday afternoon in most Fortune 500 HR departments. Here's how it actually works. The system watches for patterns. You used to join the 9 a.m. standup with your camera on. Now it's off. You stopped replying to Slack threads after 6 p.m. Your calendar has three "focus time" blocks that suspiciously match the length of a job interview. Individually, none of that means anything. Stacked together across six months, the algorithm flags you as a 78% attrition risk. Your manager gets an alert. The "check-in" happens. The pitch to employers is simple: replacing a mid-level employee costs six to nine months of their salary. If software can predict who's about to walk, a company can intervene—offer a raise, a transfer, a pep talk—before the resignation letter hits the inbox. In theory, that's good for everyone. In practice? It gets weird fast. One HR consultant told me about a client who ran predictive scores on a sales team and discovered their top performer was a "flight risk." They gave her a retention bonus. She took it, stayed eight more months, then left anyway. Another company used the same tool to identify "low performers" and laid off 40 people. Six months later, they rehired 12 of them as contractors at higher rates. The algorithm nailed the "who," but completely missed the "why." The bigger problem is what these systems don't see. They can't measure the employee who's quietly miserable because her manager takes credit for her work. They can't score the team that's about to quit together because the new return-to-office policy added two hours to everyone's commute. They can't predict the moment you decide you've had enough—only the behavioral breadcrumbs you leave behind. And here's the part that should make you uneasy: you probably agreed to this. Buried in the employee handbook you skimmed during onboarding, or the IT policy you clicked "accept" on without reading, is language about "workforce analytics" and "predictive modeling." Courts have mostly sided with employers so far. Your keystrokes, your badge data, your calendar—it's all company property. So what do you do? First, know the signs. If your company suddenly rolls out a "wellness survey" that asks about your career goals and stress levels, read the fine print. If your manager starts having unusually specific conversations about your future, someone's been reading a dashboard. Second, don't panic. Being flagged isn't a firing sentence—it's often the opposite. It's the company saying, "We'd rather keep you." Use that leverage. But also understand the trade-off. The same system that predicts your departure can predict your replacement. When you're reduced to a score, your loyalty becomes a data point—and so does your silence. **The bottom line:** Predictive intelligence isn't evil, but it's not your friend either. It's a tool, and tools serve whoever holds them. Right now, that's your employer. Until the laws catch up—and they're years behind—the best defense is knowing you're being watched, and deciding what you want them to see.