Latest / Hiring & Firing with Fexingo: Recruitment, Performance, and Workforce Decisions / How Companies Stop Paying For Ghost Employees
Transcript
- Lucas: Let’s talk about the most expensive person on your company’s balance sheet who doesn’t actually exist. We’ve spent years discussing how to hire better people, but we rarely discuss how to fire people who never showed up. In September of twenty twenty-six, ghost employees have become a silent tax on corporate profits, costing organizations an estimated one point five percent of their total payroll annually. Luna: That number feels high, Lucas. Are we talking about obvious fraud, like a manager pocketing checks for a fictional relative, or something more structural? Lucas: It used to be just the classic relative scheme. But today, it’s often more sophisticated. It involves terminated employees whose records weren’t purged from the active payroll system, or contractors who transitioned to full-time roles but kept getting paid under their old vendor codes. The key insight here is that this isn’t always malicious intent from the start; sometimes it’s just administrative decay. Luna: So the fraud is accidental at first, which makes it harder to catch because no one is looking for a crime scene. Lucas: Exactly. And that’s where our case study comes in. There was a mid-sized logistics company in Ohio that realized their warehouse overhead was climbing while productivity stayed flat. They hired an external forensic auditor, and what they found was a three hundred thousand dollar hole left by five ghost employees. Four of them had been terminated for cause over the previous eighteen months, but their bank accounts were never closed in the payroll portal. Luna: Wait, so they kept getting paid after being fired? That sounds like a basic HR failure, not necessarily fraud. Lucas: It started as a failure. But here is where it gets interesting. One of those terminated employees was actually still accessing the building. She had kept her badge. So she was physically present, showing up to work, but she wasn’t on the official roster. The company was paying her through a shell contractor agency because the original hiring manager had moved departments and forgot to reconcile the contracts. It took them six months to notice because the invoice amounts were small enough to fall below the automated review threshold. Luna: That is terrifying. You have someone walking the floor, doing work, but the books say she’s gone. How did the auditor even spot her? Lucas: They didn’t spot her through the payroll data. They spotted her through security camera metadata. The auditor ran a script comparing badge swipe timestamps against payroll check dates. He found a discrepancy in the shift patterns. Someone was swiping in during hours that didn’t match any active employee’s schedule. Once he isolated that badge ID, he traced it back to a former employee who had technically resigned but hadn’t returned her hardware. Luna: So the solution wasn’t better accounting, it was cross-referencing physical access data with financial data. That seems like a low-hanging fruit for most companies right now. Lucas: It should be. But most companies keep their HRIS, their payroll provider, and their security access control system in completely different silos. The HR director talks to finance, but nobody talks to facilities. By the time someone suggests connecting the databases, the leak has been running for years. I spoke with a chief financial officer last week who told me that his company saves two million dollars a year just by automating the deactivation of badges the moment an offboarding process starts. Luna: Two million dollars. That’s a huge ROI for a simple integration project. But does integrating those systems create privacy risks? If you’re tracking every badge swipe, aren’t you inviting surveillance complaints? Lucas: That is the ethical tightrope. You have to be careful not to treat all employees like suspects. The goal isn’t to watch everyone; it’s to protect the integrity of the compensation structure. The best firms use aggregate data for security trends and only drill down into individual anomalies when there is a financial mismatch. It’s about detecting the outlier, not monitoring the routine. Luna: Right. You want to find the needle, not burn the whole haystack. But let’s go back to the Ohio logistics firm. Did they recover that money? Lucas: They recovered about eighty percent. The rest was lost to fees or deemed too costly to pursue legally since some of the payments went to third-party agencies. But the real win wasn’t the cash recovery. It was the policy overhaul. They implemented a mandatory quarterly reconciliation between HR status updates and payroll inputs. Now, if an employee’s status changes to inactive, the payroll system automatically freezes their next deposit until a manager manually overrides it. Luna: That manual override step is smart. It forces human accountability. You can’t have an algorithm just deciding to stop payment without a supervisor signing off. Lucas: Precisely. It puts the burden of proof on the manager. If they don’t sign off, the money stays in the company account. It sounds bureaucratic, but bureaucracy is exactly what stops leaks. When we think about efficiency, we usually think about speeding things up. But in payroll, speed is the enemy. Slowing down the termination process by forty-eight hours could save you millions. Luna: I like that framing. Speed kills in payroll. But what about the psychological impact on the team? If you announce that you’re auditing everyone’s attendance to catch ghosts, morale tanks. Lucas: You’re absolutely right. You never frame it as an anti-fraud initiative. You frame it as a compliance update or a system migration. The Ohio company told their staff they were upgrading their timekeeping software to support new remote work policies. No one panicked. They just got stricter about clocking in and out. The ghosts vanished because the system became harder to exploit, not because anyone was accused. Luna: So it’s a technical fix disguised as an operational upgrade. Clever. But is this a problem mostly for large enterprises, or can small businesses fall victim to this too? Lucas: Small businesses are actually more vulnerable because they lack dedicated audit teams. A small marketing agency might have ten employees. If one person leaves and their direct deposit isn’t cancelled, the owner might not notice for months because they’re wearing ten other hats. The absolute loss is smaller, maybe fifty thousand dollars, but that’s half the annual bonus pool for a small firm. The percentage hit is devastating. Luna: Yeah, losing half your bonus pool hurts way more than losing one percent of revenue. What’s the first step a small business owner should take tonight? Lucas: Pull the active payroll list and compare it against your email directory. Look for names that are in payroll but not in the company directory. Or vice versa. If someone is in the directory but not getting paid, they might be a ghost contractor who should be an employee, or vice versa. It takes twenty minutes and exposes half the problems immediately. Luna: Twenty minutes to potentially save thousands. That’s the kind of actionable advice listeners actually need. Before we dive deeper into the tech stack required for this, I want to pause for a second. If these conversations have moved your work forward in some small way, consider supporting the show directly. Honest, ad-free business analysis costs money to produce. If today was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. It helps us keep digging into these messy operational details. Lucas: Appreciate it. It really does help us stay independent and focus on the long game rather than chasing clicks. Luna: Thanks for listening. Anyway, back to the tech stack. Lucas, you mentioned the Ohio firm integrated security data with payroll. Do they use a specialized vendor for that? Lucas: They didn’t buy a new vendor. They used a middleware tool that connects their existing HRIS, like Workday or BambooHR, to their physical security system, like Lenel or Genetec. The middleware runs a daily job. It looks for mismatches. If it finds one, it flags it for the HR operations team. It’s not ai driven prediction; it’s just deterministic logic. If record A says inactive, and record B says active payment, flag error. Luna: Deterministic logic is much easier to sell to a CFO than AI. You can explain exactly why it caught the ghost. With AI, you get black box results that people distrust. Lucas: Right. And trust is everything in payroll. If your employees don’t trust that their pay is accurate and secure, you have a culture problem far bigger than ghost employees. The audit is a hygiene factor. It’s table stakes for a mature organization. Luna: Table stakes. I love that. But let’s play devil’s advocate. What if the ghost employee is a legitimate person, but they are working remotely and haven’t updated their address? Is that still a ghost? Lucas: No, that’s just outdated information. The term ghost employee specifically refers to someone who is either deceased, terminated, or entirely fictitious. Remote workers are active employees. The confusion arises because remote work makes it harder to verify physical presence. But you don’t need to verify presence to verify employment status. You just need to verify that the person on the payroll is authorized to be there. Luna: Got it. So the metric is authorization, not location. That simplifies the definition. But what about international subsidiaries? If a US parent company hires a subsidiary in Vietnam, can they still pull those badge logs? Lucas: That is a massive blind spot. Many multinationals lose millions in overseas entities because the parent company assumes the local subsidiary has its own controls. Often, they don’t. The local manager might be using a spreadsheet. The audit has to go global. The Ohio firm actually had a similar issue in their Mexican distribution center, where they found three ghost workers on the local payroll that the US headquarters couldn’t see. Luna: Three workers in Mexico. That adds up fast. It sounds like centralized visibility is the only real cure. You can’t fix what you can’t see across borders. Lucas: Exactly. Centralization isn’t just about convenience; it’s about risk mitigation. If you have a fragmented global workforce, you are leaving money on the table. Every disconnected system is a potential leak. The trend in twenty twenty-six is toward unified identity platforms. These platforms manage your digital identity, your physical access, and your financial records in one place. When you terminate a user, you flip one switch, and the check stops, the badge dies, and the email locks. Luna: One switch. That’s the holy grail. It removes the human error element entirely. But implementing that is expensive and disruptive. Is it worth the pain for a company that hasn’t had a ghost employee problem yet? Lucas: It’s cheaper than the alternative. Think of it as insurance. You hope you never need it, but when you do, you’re glad you bought it. Plus, the efficiency gains from automation often pay for the platform within twelve months. You stop spending hours reconciling spreadsheets manually. Your HR team can focus on strategy instead of data entry. Luna: Shifting from data entry to strategy. That’s the real value proposition. Okay, final question on this topic. What’s the biggest mistake companies make when they finally decide to clean house? Lucas: They panic. They try to fire everyone suspected of being a ghost simultaneously. That causes chaos. It shuts down operations. The right move is a surgical strike. Identify the top three highest-value anomalies, investigate them quietly, resolve them, and then roll out the systemic fix. Don’t shake the tree. Prune the dead branches. Luna: Surgical strike. I’m writing that down. Prune the dead branches. It’s a great metaphor for workforce management. Thanks for breaking this down, Lucas. It’s scary how easy it is to lose money silently. Lucas: It is. But now that you know where to look, you can’t unsee it. The question is, will you fix the leak before it drains the tank?