Details in this case study have been anonymized to protect client confidentiality. The financial figures, timeline, and detection mechanism are accurate.
The Number That Should Have Been Fine
The CEO of a multi-location retail business had been running his company for over a decade. He knew his numbers. He knew his inventory cycles, his seasonal patterns, his cash position. He had built the kind of operational intuition that comes from years of watching the same rhythms repeat — the post-holiday restocking, the spring build, the summer plateau. He had been through tight months before. He knew what normal looked like.
Which is why, when Finteligence flagged an anomaly in his cash runway data, his first instinct was that something had been miscategorized. The number on the screen didn't match his mental model of where the business should be. His runway had dropped — not gradually, not in a way that tracked with any seasonal pattern he recognized, but sharply, in a compressed window, in a way that put him below thirty days of operating cash.
He was right that something was wrong. He was wrong about what it was.
How the Business Managed Inventory
To understand what happened, it helps to understand how the business was set up. Like many multi-location retail operations, inventory purchasing ran through a dedicated card. The card was managed by the employee responsible for purchasing. It was set up to auto-pay when the balance reached a threshold — a common and sensible arrangement that kept the account current without requiring manual intervention on every cycle.
The system worked well. It was efficient, it was automated, and it had been running without incident for years. The purchasing employee knew the vendors, knew the order cadence, and knew the seasonal patterns. After the holiday season, the standing order was to replenish — a larger-than-normal order to replace the inventory that had moved during the peak period. This was expected. This was planned.
What was not planned was that the order went through twice.
Not through fraud. Not through negligence in any meaningful sense. The employee had placed the post-holiday replenishment order and, through a process error that is entirely ordinary in the context of high-volume purchasing, the order was submitted a second time. Two identical orders. Two identical charges to the card. The auto-pay threshold triggered on both. The cash left the account before anyone had a reason to look.
What Finteligence Flagged
The Finteligence monitoring layer identified the anomaly through a combination of two signals. The first was the cash runway drop itself — a sharp, compressed decline that did not fit the business's historical pattern for this point in the seasonal cycle. Post-holiday restocking always produced a cash draw, but the magnitude and timing of this one fell outside the expected range.
The second signal was the inventory purchase pattern. The charge that had triggered the runway decline was categorized as inventory — consistent with the post-holiday replenishment cadence — but the amount was approximately double what the historical pattern indicated for this period and this vendor. A single large order would have been unremarkable. Two identical charges, processed in close succession, from the same vendor, for the same product mix, was a pattern that did not match any prior cycle in the data.
The Finteligence team reviewed the flags and contacted the client. The conversation was straightforward: the cash runway had dropped below thirty days, the driver appeared to be an inventory purchase that didn't fit the normal pattern, and it warranted a look before the next reporting cycle closed.
The CEO pulled the card statement that afternoon.
The Call the Next Morning
He called back the following morning. He had found the duplicate order immediately once he knew where to look. The purchasing employee confirmed it — a process error during the post-holiday replenishment, the kind of thing that happens in high-volume operations and is usually caught before it clears. This one had not been caught, because the auto-pay mechanism had processed both charges before anyone reviewed the statement.
The merchandise had shipped but had not yet arrived at the warehouse. The CEO called the vendor the same day. Because the error was identified within the delivery window, the vendor agreed to accept the return of the duplicate order and issue a full refund. The cash was recovered. The runway stabilized.
The entire sequence — from the Finteligence flag to the vendor call to the confirmed refund — took less than seventy-two hours.
What Would Have Happened Otherwise
This is the part of the story that matters most, and it is worth being precise about it.
Under the business's previous reporting structure — monthly closes, delivered by the advisory firm, reviewed at the end of the period — the duplicate order would not have appeared as a problem until the month-end report. By that point, the merchandise would have arrived, been received into inventory, and been integrated into the stock. A return would have been significantly more complicated, potentially subject to restocking fees, and dependent on vendor goodwill that is harder to secure weeks after the fact rather than days.
More critically: the cash would have been gone for the full duration of the reporting cycle. With a runway below thirty days at the point of detection, the business was operating with a margin of error that left almost no room for any additional disruption — a delayed receivable, an unexpected expense, a slower-than-projected sales week. Any one of those things, in combination with the compressed runway, could have created a cash crisis that the business would not have been able to absorb.
The CEO said it plainly in the follow-up conversation: "By the time we caught it through normal reporting, we'd have been out of operating cash."
He was not being dramatic. He was describing the arithmetic.
The Case for Continuous Monitoring
The fraud cases — the construction foreman, the split-check scheme, the $68,000 over four years — are compelling because they involve intentional wrongdoing. They are the cases that get shared at industry conferences and cited in advisory firm marketing. They are real, and they matter.
But the retail runway case is, in some ways, a more important story. Because it did not involve fraud. It did not involve a bad actor or a deliberate scheme. It involved a process error — the kind of ordinary operational mistake that happens in every business, at every size, in every industry. The purchasing employee was not careless. The auto-pay system was not poorly designed. The vendor relationship was not problematic. Everything was working as intended, and a single process error nearly ended the business.
The question that case raises is not "how do you catch fraud?" The question is: "how much time passes between when something goes wrong and when you find out about it?" In a business running on monthly reporting, the answer is up to sixty days. In a business running on Finteligence continuous monitoring, the answer is hours.
For a business with thirty days of runway, the difference between those two answers is the difference between a recoverable error and an unrecoverable one.
That is what the FinTel category provides. Not a guarantee that nothing will go wrong — nothing provides that. A guarantee that when something does go wrong, you find out while you can still do something about it. The real-time FinTel era is here. The question is whether your advisory relationship has been structured to deliver it.
Melissa Lewis is the founder and CEO of Sentinel Intelligence Corp., the company behind Finteligence — a continuous financial intelligence platform delivered exclusively through advisory partnerships with CPA and CFO firms. If you are a CPA or fractional CFO interested in offering Finteligence to your clients, visit the advisory partner page.