Latest / Future of Work Tech with Fexingo: Remote Tools, AI Productivity, and Workplace Software / Why Ghost Employees Are Costing Companies Billions
Transcript
- Lucas: So there's this manufacturing company in Ohio — about 800 employees, family-run, been around since the 1970s. They do precision metal stamping for auto parts. Solid business. Luna: Sounds boring but stable. What happened? Lucas: A routine payroll audit flagged something odd. They had 847 people on payroll but only 800 active badges swiping into the plant. That's a 47-person gap. Turns out, those 47 were ghost employees — people on the books who didn't actually work there. Luna: Wait — how do you get a fake person onto payroll? Don't you need a Social Security number, a bank account, all that? Lucas: You'd think. But in practice, it's surprisingly easy, especially in companies with decentralized hiring. In this case, a mid-level HR manager had been adding fake entries using real Social Security numbers from a data breach. The direct deposits went to prepaid debit cards. Luna: That's terrifying. And this was going on for how long? Lucas: Three years. Total loss: $2.3 million. And this isn't a tiny anomaly. The Association of Certified Fraud Examiners estimates that payroll fraud, including ghost employees, costs organizations an average of 5 percent of annual payroll. For a company like that, that's roughly $4 million a year. Luna: So 5 percent — just quietly leaking out. And I'm guessing remote work has made this worse? Lucas: Significantly. Pre-pandemic, ghost employees were mostly a brick and mortar problem — you'd notice if someone never showed up to the office. But now, with fully remote teams, you can have a person on payroll for months without anyone ever seeing them. In fact, the Department of Labor estimates that ghost employees cost U.S. businesses about $150 billion annually. Luna: That number is staggering. So how do companies catch this? Is it just audits? Lucas: Traditional audits help, but they're usually backward-looking — you find the fraud after it's happened. What's changing is ai powered payroll verification. Startups like PayrollGuard and WorkNumber are using biometric cross-referencing. For example, they match time-sheet data with badge swipes, IP addresses, even facial recognition on check-in. Luna: Facial recognition for payroll — that feels invasive. What about privacy pushback? Lucas: It's a real tension. But the companies that deploy it argue that the alternative is leaving the door open for fraud. PayrollGuard, for instance, uses a zero-knowledge proof system — they verify identity without storing the actual biometric data. So the privacy risk is lower than you'd think. Luna: I want to come back to that privacy angle. But first — how common is it that the ghost employee is an insider? Like, is it always an HR person cooking the books? Lucas: Interesting question. The ACRE report says about 60 percent of ghost employee cases involve an employee in payroll or HR. But the other 40 percent? That's managers adding friends or relatives, or even contractors who get added to payroll and never terminated. Luna: So it's not just one bad actor — it's a systemic issue. And I imagine in smaller companies, there's less oversight. Lucas: Exactly. A 50-person startup might not have a dedicated HR auditor. They rely on the founder or the CFO to eyeball the payroll register. But when you're growing fast and hiring remote, it's easy for a fake entry to slip through. I talked to a founder of a 120-person SaaS company who discovered four ghost employees after six months — one was a former intern who never got removed. Luna: You know, this conversation is exactly why we keep this show ad-free — because digging into these specific cases takes time and resources. If today's episode gave you something useful, the way we keep going is listener support. You can head to buy me a coffee dot com slash fexingo. It's a small gesture that keeps us independent. Lucas: Yeah, absolutely. That support directly funds the research and the conversations we have here. So if you found value in understanding how ghost employees work, that link is a great way to ensure we keep uncovering these stories. Lucas: Now, back to the detection side. Beyond biometrics, what else are companies doing? One approach is behavioral analytics. WorkNumber's platform looks at patterns — if an employee's timesheet is always exactly 40 hours, never takes vacation, and their direct deposit has never changed, that's a red flag. Luna: So the fraudsters are almost too perfect. They avoid anything that would trigger a manual review. Lucas: Exactly. A real employee might have occasional overtime, a vacation request, a change in bank account. Ghost employees are often too consistent. And the AI catches that. Luna: What about the role of regulation? Is the government doing anything to mandate these checks? Lucas: The Department of Labor has been pushing for more stringent verification, especially for companies with government contracts. But there's no federal mandate yet. Some states are ahead — California, for instance, requires E-Verify for all new hires, which cross-references with Social Security. But that doesn't catch ghost employees who use real SSNs from breaches. Luna: So it's still largely on companies to self-police. And the cost of not doing it is huge — both financially and reputationally. Lucas: Right. And the reputational hit can be brutal. In 2023, a well-known retail chain had a ghost employee scandal where a store manager added 12 fake employees over two years. The company had to restate earnings and the stock dropped 8 percent. Luna: Wow. So this isn't just a back-office nuisance — it's a material risk to shareholders. Lucas: Exactly. And the more remote work becomes permanent, the bigger the attack surface. Companies need to treat payroll as a security function, not just an accounting function. Luna: So what's the first step for a company that wants to check if they have ghost employees? Lucas: Start with a simple cross-check: pull the payroll register and compare it against active employee lists from HR, IT, and facilities. Look for duplicates, missing contact info, or employees who haven't used their corporate email in months. Then run a match against time-tracking data. If you see someone with full-time pay but zero time entries, that's a red flag. Luna: And if they find something, should they go public or handle it quietly? Lucas: Most companies handle it quietly — terminate the perpetrator, implement new controls, and possibly report to law enforcement if it's a large sum. But transparency with investors is usually required if it's material. The SEC has started asking about payroll fraud in audits. Luna: So the era of 'trust but verify' is really becoming 'verify, then verify again'. Lucas: Exactly. And with AI tools getting cheaper, the cost of verification is dropping fast. PayrollGuard charges about $2 per employee per month. For a 1,000-person company, that's $24,000 a year — a fraction of the potential loss. Luna: Seems like a no-brainer. But I wonder — are there any false positive risks? Like, could a legitimate employee get flagged and then face scrutiny? Lucas: It happens. The algorithms can sometimes flag employees who genuinely have very regular schedules or who don't take vacation. That's why human review is still important. The best systems flag anomalies, not condemn them. They create a dashboard for HR to investigate. Luna: So it's a tool, not a solution. You still need good HR practices. Lucas: Exactly. And that's the key takeaway — technology is an enabler, but culture and process are the foundation. If a company has a culture of 'we trust everyone,' they're vulnerable. A little paranoia in payroll goes a long way. Luna: Alright, so to recap: ghost employees are a real and growing problem, especially with remote work. They cost companies billions. AI and biometric tools can help, but they need to be paired with good audits and a healthy skepticism. Lucas: That's the story. And if there's one number to remember, it's 5 percent — the average leak in payroll. That's a huge opportunity for companies to plug a hole they might not even know they have. Luna: And for listeners, if you're in HR or finance at your company, maybe worth a quick check this week. Lucas: Absolutely. Thanks, Luna. Luna: Thanks, Lucas.