The Checkout System That Almost Cost Miriam Her Purchase

The situation

Miriam had eleven days left before the financing commitment attached to her offer expired, and the gas station's point-of-sale system still would not close a single shift without crashing partway through the reconciliation. She had worked the counter at the station for six years, ringing through fuel, cigarettes, lottery tickets and the trays of bread and pastries baked fresh each morning at the attached bakery counter Dov had added a decade earlier. When Dov decided to retire, he offered to sell the business to Miriam rather than list it, and she agreed to buy it for a price in the low six figures, financed mostly through a small business loan she had spent months arranging.

The deal looked simple on paper. Miriam knew the business from the inside, Dov wanted a clean exit, and the two of them had a good working relationship built over years of early mornings and slow winter afternoons. The complication was the point-of-sale system. Three months before the sale was supposed to close, Dov had switched providers, moving fuel sales, bakery sales and loyalty tracking onto a new platform after the old vendor announced it was shutting down support. The migration had been rushed, handled by a local consultant Dov brought in on his accountant's recommendation, and it had never fully worked. Shift reports did not balance. Loyalty credits from the old system had not transferred. Nobody could say with confidence how much inventory the bakery counter actually had on hand.

Dov's accountant, Raymond, had reviewed the sale numbers before the offer went to Miriam and had signed off on the financials as sound, treating the point-of-sale problems as a technical inconvenience rather than something that touched the value of the business. Miriam's lender was less relaxed. Days before the financing deadline, the bank's underwriter flagged the unreliable sales data as a reason to hold back part of the loan until the numbers could be trusted, and the clock on Miriam's financing condition kept running regardless.

By the time Miriam came to us, she was eleven days from losing her financing commitment, holding a business she desperately wanted, built on sales records neither she nor Dov could fully vouch for, with the person meant to catch this kind of problem having already told everyone it was fine.

Why this was harder than it looked

The obvious fix looked like a straightforward extension: ask the lender for more time until the point-of-sale data could be reconciled. In practice, extending a financing condition close to its deadline is rarely simple, because every day the condition stays open is a day the seller can walk to another buyer, and Dov, however fond he was of Miriam, was retiring on a timeline of his own and had already fielded one unsolicited inquiry about the station from someone else.

The deeper problem was what the broken migration had actually cost. Reconstructing accurate shift totals meant going back through paper tapes, bank deposit records and supplier invoices for the three months since the new system went live, because the software itself could not be trusted to reconcile on its own. That work took time neither side had budgeted, and it kept surfacing small discrepancies that were individually minor but collectively made it hard to say what the bakery counter and fuel operation were really earning in a typical month.

Raymond's earlier sign-off complicated things further. Because Dov's accountant had already told him the numbers were fine, Dov was reluctant to accept that there was a real gap, and any renegotiation of price risked reading as an accusation against someone he trusted. Miriam, for her part, could not simply walk away from the deal without losing the loan work she had already put in and the only realistic path she had to owning a business she had spent years helping to run.

Raising the issue at all carried its own risk. If Miriam pushed too hard on the reconciliation and framed it as Dov's accountant having failed him, she risked damaging a relationship she needed to stay intact through closing and beyond, since Dov's cooperation on the transition, introductions to the fuel distributor and a smooth handover of day-to-day operations mattered just as much as the price on paper. The approach had to surface the real gap in the numbers without turning it into a referendum on Raymond's competence.

There was also a practical wrinkle in the loyalty program. Regular customers had accumulated credits on the old system that had not carried over cleanly, and whoever ended up owning the station would be the one fielding complaints from customers who believed they were owed free coffee or bread that no longer showed up anywhere. Nobody had put a number on that exposure, and it needed one before the deal could reasonably close.

None of these problems, on their own, were large enough to blow up the purchase. Together, on a financing deadline measured in days, they were enough to put the whole transaction at risk.

There was also a timing problem layered underneath all of it. The lender's underwriter had not asked for a full audit of three months of transactions; she had asked for enough confidence in the numbers to lend against them, which meant the work had to be fast as well as credible. A slow, thorough reconciliation that produced perfect numbers two months after the financing deadline had already lapsed would have solved nothing. Whatever path forward existed had to move at the speed of a deadline that was already counting down, not at the speed a careful bookkeeping review would normally take.

What we did

1. Secured a short, defined extension. Rather than asking the lender for an open-ended delay, we asked for a specific ten-business-day extension tied to a concrete deliverable: a reconciled set of shift and inventory records prepared by an independent bookkeeper. A narrow, deliverable-based extension is easier for a lender to approve than a vague request for more time, and it gave Dov a firm date to hold onto instead of an uncertain drift that might have pushed him toward the other interested buyer.
2. Brought in an independent bookkeeper to reconcile three months of records. Rather than relying on the point-of-sale system or on Raymond's earlier review, we arranged for a bookkeeper with no connection to either side to work back through bank deposits, supplier invoices and the paper tapes kept as a backup during the migration. That reconciliation gave both Miriam and Dov a set of numbers neither of them had reason to distrust, which turned out to matter more than the numbers themselves.
3. Put a figure on the loyalty liability. We asked the bookkeeper to estimate, from the surviving records of the old system, roughly how much unredeemed loyalty credit was outstanding when the migration happened. The estimate came in modest, in the low thousands, but it mattered because an unquantified liability is much harder to negotiate around than a number both sides can argue over and eventually accept.
4. Renegotiated price against the reconciled numbers, not the estimate. Once the independent numbers were in hand, we used the gap between what Dov's original figures implied and what the reconciliation actually showed to negotiate a modest reduction in the purchase price, tied specifically to the shortfall the records revealed rather than to any suggestion Dov had acted in bad faith. Framing it around the numbers, not blame, kept Dov willing to keep negotiating instead of digging in.
5. Structured a holdback instead of a flat discount. Because a handful of reconciliation items were still unresolved even after the bookkeeper's work, we proposed holding back a portion of the purchase price in escrow for a set period after closing, released to Dov once the final numbers settled. That let the deal close on schedule while protecting Miriam against a discrepancy that might still surface once she was running the till herself.
6. Addressed the loyalty program directly in the purchase agreement. We built a clause into the agreement setting out exactly how outstanding loyalty credits would be honoured after closing and capping Miriam's exposure at the estimated figure, rather than leaving the obligation open-ended. That cap mattered because loyalty credits tend to surface gradually over months rather than all at once, and without a ceiling, a wave of customers redeeming old credits could have become a cost Miriam had never priced into her financing.
7. Went back to the lender with a complete package. With the reconciliation, the loyalty estimate and the revised price in hand, we worked with Miriam's lender directly to show that the financing condition's underlying concern, unreliable sales data, had been addressed with real numbers rather than assurances, which let the loan proceed on its original terms within the extended window.

The outcome

The sale closed twelve days after the original financing deadline, inside the extension we negotiated, at a price reduced by an amount in the mid five figures from the original offer, with a further portion held back in escrow for ninety days while the last reconciliation items settled. Miriam did not get the business at the price she first agreed to, and she did not get every dollar of the loyalty exposure covered outright, but she got a deal built on numbers she could stand behind rather than ones she had simply been asked to trust.

Dov accepted the reduced price without treating it as a verdict on Raymond's work, in part because the negotiation stayed anchored to the independent reconciliation rather than to any claim that the original accountant had been negligent. That distinction mattered to him, and it kept the sale from turning into a dispute neither side had the appetite for so close to his retirement.

The escrow released in full a little over three months after closing, once the last few discrepancies in the migration records were traced to timing differences rather than missing money. Miriam ended up running the station with a point-of-sale system she had already stress-tested through the reconciliation process, and a loyalty program with a known, capped cost instead of an open question hanging over the till.

The compromise was not a clean win for either side. Dov left with less than he expected, and Miriam paid a price that still felt tight against her financing. What both of them avoided was a collapsed sale, a lender that walked away, or a dispute over Raymond's earlier sign-off that would have cost far more than the discount ultimately did.

A year later, Miriam's shift reports balanced without incident, and the loyalty program had settled into a routine cost well inside the cap the agreement had set. Dov, for his part, kept working with Raymond on his personal finances after the sale closed, the earlier miss treated as a lesson rather than a betrayal. Neither outcome was guaranteed when Miriam first walked in with eleven days on the clock.

What you can learn from this