Financial Accounting and Bookkeeping Case Studies
Finance5 Case StudiesUpdated Sep 2026
These five cases come from ordinary bookkeeping and close work, where numbers look settled until someone tries to tie them together. Each begins with a variance, a forecast that stopped matching the ledger, a flagged transaction, a credit request, or an audit finding, and each asks what the record shows. All five turn on the same four things: transaction records, ledger integrity, approval controls, and close deadlines. The organizations are fictional, but month-end pressure will look familiar to anyone who has sat through a close review.
Case Study 1
The Unmatched Receipts Behind a Cold Storage Variance
Kestrel Bay Seafood Distributors, a regional wholesaler with three cold storage sites, about 140 staff, and a single shared ERP ledger for inventory and payables.
The situation
Kestrel Bay closes its books on the fourth business day after month end. In the most recent close, the inventory subledger valued received goods at 412,850, while the general ledger control account for inventory stood 11,460 lower. The gap was not spread evenly. Most of it sat in receipts from one Harbor Street dock, where receiving staff had entered three weeks of deliveries in a batch on the last afternoon. Separately, the operating bank reconciliation carried a 2,315 debit that no payable had claimed. The controller, Marisol, had four business days to explain both before the board pack went out.
The challenge
The variance could come from timing, a posting error, a duplicate receipt, or a supplier invoice approved against the wrong purchase order. Each explanation predicts a different pattern in the transaction records. Marisol needed to know which one fit before choosing a correction, because a plug entry to the control account would make the close tie out while hiding the cause. The batch entries also carried approval timestamps that came after the subledger had already been locked for the period.
Analysis
Teams grounded in Financial Accounting and Bookkeeping usually start with the ledger rather than the explanation. The controller pulled every receipt line with a posting date in the final two weeks, then matched each one against the goods receipt number, the supplier invoice, and the purchase order. The pattern mattered more than any single line. Duplicate receipts clustered on suppliers paid in the same batch run, and the bank debit traced to a payment released twice when a remittance file was re-sent. Ledger integrity here meant confirming that the control account had been posted only from the subledger journal, with no manual entries sitting on top of it.
The approval gap was the real finding. The batch receipts were released by a single supervisor who also held rights to post journal adjustments, which meant the same person could create a variance and clear it. Under the close calendar, a receipt posted after the subledger lock should have required a second approver and a dated exception note, and neither existed. Marisol did not blame the clerk, who had followed a habit the system allowed. She treated the missing segregation as the control failure and set a rule that no inventory variance could be cleared by the person who created the underlying entry.
- 1Subledger and ledger comparedTotals differ by 11,460 at the close lock point.
- 2Receipts matched to documentsGoods receipt, invoice, and purchase order checked line by line.
- 3Duplicates and timing isolatedThree double receipts and one cutoff item explain the gap.
- 4Bank debit tracedRemittance file resent, releasing one payment twice.
- 5Controls redesignedPosting rights split and a second approver added.
Actions taken
- Reverse the duplicate receipts and the double-released bank debit, attaching the original source document to each correction so every adjustment can be traced.
- Remove journal posting rights from the warehouse supervisor role and require a second approver for any manual entry to the inventory control account above 1,000.
- Set a receiving cutoff so deliveries logged after the third business day roll into the next period with a documented accrual, not a backdated receipt.
- Run a daily tie-out of the subledger to the control account during the last five business days, with reconciling items signed by someone outside the posting team.
Results
Within the same close cycle, the inventory control account tied to the subledger without a plug entry. The 11,460 variance resolved into three duplicate receipts and one timing item, and the 2,315 bank debit was recovered from the supplier through a credit memo. The board pack went out on the fourth business day. Two months later the reconciling items list had shrunk to a handful of timing differences, each with a named owner and a clearance date. The warehouse supervisor now sends receiving exceptions to the controller every day rather than at close.
Questions for discussion
- Which of the four candidate explanations would have left a different pattern in the receipt data, and how would you test for it before posting anything?
- Who in your own close process can both create an entry and clear the variance it causes, and what would it take to separate those roles?
- When a reconciliation ties only because of a plug, what does that tell you about the controls that were supposed to catch the difference earlier?
Case Study 2
Holding the Cash Forecast When Spring Sales Stall
Ironbark Outdoor Gear Company, a mid-sized maker and online retailer of camping equipment with two warehouses and a strongly seasonal sales pattern.
The situation
Ironbark planned its spring season around a revenue target of 6,480,000 for the quarter. Six weeks in, orders were running about 14 percent behind that pace, and the shortfall was concentrated in tents and sleeping systems, the two categories with the largest supplier commitments. Purchase orders for the season had been placed in the prior quarter, and 2,300,000 of them fell due within ninety days. The finance lead, Dariusz, rebuilt the cash forecast weekly from the sales ledger and the open payables file. The revolving facility carried a minimum liquidity covenant, and the latest forecast showed headroom narrowing toward it by the end of the quarter.
The challenge
The shortfall could be a timing effect from a late marketing push, a structural change in demand, or a pricing error after a promotion ended. Each reading implies a different response. Cutting purchases hurts if demand recovers, while waiting risks a covenant breach if it does not. Dariusz also had to decide which assumptions in the model could be defended in front of the lender and which were simply carried forward from last season.
Analysis
The first step was to separate price from volume. Dariusz pulled order-level transactions and compared units shipped, average selling price, and discount codes against the plan, rather than trusting the revenue total. The unit gap explained most of the shortfall, but the discount ledger showed a promotion code that had been applied to a wholesale account at retail-channel terms, which inflated the price variance in the opposite direction. He corrected the mapping in the ledger rather than adjusting the forecast, since the forecast was only as good as the postings beneath it. Each scenario was then built from cleaned transaction data, with the source of every input recorded.
Scenarios came next. The base case assumed the sales pace recovered by the middle of the quarter, a downside case held the current pace through the end of the period, and a severe case cut the pace further while assuming supplier payments could not be deferred. The test that mattered was the timing of the 2,300,000 payments against the weekly collections curve. Under the existing approval rules, any supplier deferral above 150,000 needed the chief executive's sign off, and the close calendar put the next lender compliance certificate about six weeks out. That gave the company a window to renegotiate terms before the covenant test became a fact rather than a forecast.
- Six weeks inOrders running about 14 percent under plan, concentrated in tents and sleeping systems.
- Seven weeks inLedger discount mapping corrected; apparent price gap narrowed by about a third.
- Eight weeks inPurchase holds placed and supplier deferral talks opened with largest vendors.
- Within ten weeksLender briefed on three scenarios with pre-agreed triggers and actions.
- End of quarterCovenant headroom tested against the base, downside, and severe cases.
Actions taken
- Freeze new purchase orders for the two lagging categories until weekly sell-through confirms the recovery assumption, and log each hold with its approver and date.
- Open deferral talks with the three largest suppliers, offering staggered payment dates in exchange for firm delivery windows, and document each agreed change in the payables ledger.
- Correct the wholesale discount mapping in the sales ledger and rerun the variance analysis, so the forecast starts from postings that match contract terms.
- Brief the lender early with three scenarios, the trigger point for each, and the specific actions tied to each trigger, rather than waiting for the compliance certificate.
Results
The team froze purchases in the two lagging categories and secured deferrals from two of the three largest suppliers, moving 640,000 of payments out of the next thirty days. The corrected discount mapping reduced the apparent price shortfall by roughly a third, which changed the story from a pricing problem to a volume problem concentrated in one season. Sell-through improved in the fourth week after the promotional change, and forecast headroom stayed above the covenant in the base and downside cases. The severe case still breached, which is why the lender briefing listed the purchase cut that would apply if the pace did not recover.
Questions for discussion
- Which assumption in your own cash forecast would you test first, and what transaction evidence would convince a lender that it holds?
- How should a finance lead decide between cutting purchases now and waiting for demand data that may arrive too late for the covenant test?
- What would a corrected ledger mapping change about the story told to the board, even if the total revenue figure stays the same?
Case Study 3
Deciding Whether Vendor Payments Need Escalation
Greyhollow Ledger Services, an outsourced bookkeeping and payroll firm serving about 60 small and midsize clients, with a compliance officer who reviews all escalations.
The situation
Over the last three months, the payables ledger for Sallow Creek Home Builders picked up fourteen payments to a single consulting vendor, Northmere Consulting Partners. Each one fell between 9,800 and 9,950, just under the 10,000 level at which the client's bank requires a second release. The vendor master record listed only a post office box, and no tax identification number had been entered until the sixth payment. The payments were approved by the office manager and one of the two owners, and no contract had been filed with the bookkeeping records.
The challenge
Greyhollow's engagement covered bookkeeping and payroll, not an audit or a legal review. The question was whether the pattern reflected an awkward but ordinary vendor relationship or a structure built to stay under the bank's review level, and what the firm owed its client, its own staff, and possibly an outside authority. Lucia, the senior accountant on the account, had to decide what to record, what to hold, and who to tell before the next payment run went out.
Analysis
The first task was documentation, not judgment. Lucia traced each of the fourteen payments back to a bill, a bank confirmation, and an approval record. She found that three bills carried the same invoice number with different amounts, and that two payments had no bill at all. She also checked whether entries had been adjusted after the fact. Several had been reclassified from supplies to consulting in the final week of each month, a pattern worth noting because it changes how the expense appears in the close. Nothing in the records proved intent. What they did show was that the books could not support the payments as recorded.
Exposure was the next question. The payments totaled 137,400 over the period, and the client's own policy prohibited vendor payments without a signed contract above 5,000. Greyhollow's engagement letter and its internal anti-money-laundering policy both required the firm to escalate unexplained patterns to its compliance officer rather than resolve them quietly with the client. Because the firm was not the party approving payments, its duty was to document the indicators, stop preparing entries that treated the vendor as normal, and advise the client in writing. The compliance officer, Teodor, reviewed the file before any conversation with the owners took place.
Actions taken
- Place a hold on the vendor record so no further payment can be released until a signed contract, a tax identification number, and a physical address are on file.
- Move the unsupported reclassifications and the two bill-less payments to suspense, with a memo linking each entry to the document reviewed.
- Write to the owners stating which records were missing and which entries were held, without speculating on intent, and keep a copy in the engagement file.
- Ask the compliance officer to decide in writing whether the pattern meets the firm's reporting threshold, and record that decision and its date before any further entries are posted.
Results
Teodor reviewed the file within two days and agreed with the hold. The owners produced a signed contract covering four of the fourteen payments but could not document the rest, so the unsupported amounts sat in suspense pending review. Whether the matter needed a report was decided by the compliance officer, not the bookkeeping team, and that decision was written into the file. The client moved vendor payments to a dual sign-off process, which Greyhollow now checks each month.
Questions for discussion
- Which of the three documentation gaps would you resolve first, and what would count as sufficient support for a payment that has already cleared the bank?
- When the client's own owners approve the transactions, what duty does an outside bookkeeper have to the firm's compliance function?
- How would you design a close checklist so that a vendor record without a tax identification number could not drive a payment in the first place?
Case Study 4
A Bakery Expansion Loan With a Thin Cash Cushion
Ashgrove Bakehouse Holdings, a family-owned group of three bakeries and a wholesale kitchen, with about 90 employees and steady growth over the last several years.
The situation
Ashgrove Bakehouse Holdings asked the credit committee for an 850,000 seven-year term loan to open a fourth shop and replace two deck ovens. Revenue for the last full year was 3,920,000, with operating earnings before depreciation of 610,000 and equipment debt of 310,000 already on the books. Year-end cash was 74,000, roughly two and a half weeks of payroll. A regional grocer accounted for 31 percent of revenue under a supply contract with a renewal date that fell within the loan's first year. Imani, the credit analyst, received the request with two years of monthly management accounts, though the most recent four months had not yet been closed by the bookkeeper.
The challenge
The earnings figure was healthy enough to support the loan at face value, but three things complicated it. The cash cushion was thin, the largest customer contract was short and renewable, and the latest four months were unreconciled. The committee needed to know whether repayment capacity held up under a bad season, and whether the owners' projections rested on numbers the books could support.
Analysis
Imani started with what the ledger could prove. She rebuilt the trailing twelve months from bank statements and the sales journal rather than the management summary, and the two differed by about 46,000 in revenue, mostly from wholesale invoices recorded on delivery rather than on shipment. That timing difference does not change annual earnings, but it changes the quarter-end cash picture that the projections were built on. Two months showed a pattern of late supplier payments being posted to the following period, which flattered reported margins. Until those four unclosed months were reconciled, she treated the reported earnings as a range, not a point.
The downside test used the grocer contract as the main stress point. In the reduced case, the grocer's volume fell by half at renewal, and wholesale margin tightened with it, cutting annual operating earnings to roughly 400,000. Debt service on the proposed loan would then consume most of the cash flow, leaving the cushion below one month of payroll. The owners' projection assumed the new shop reached break even within its second year, which Imani judged plausible but not yet evidenced by any closed month. Suitability pointed toward a smaller facility, with the ovens financed separately against their resale value and a drawdown tied to the lease signing and the grocer's written renewal.
- Verify books Rebuild trailing revenue and cash from bank and sales records.
- Stress test Model grocer loss, thin cash, and a slow second-year shop.
- Set conditions Tie drawdowns to the lease and a written grocer renewal.
- Review quarterly Compare closed months against the projection and release tranches.
Actions taken
- Require the four unclosed months to be reconciled and closed before committee approval, with the bookkeeper certifying the year-end cash balance against bank confirmations.
- Restructure the request into a 600,000 term loan for the shop and a separate 250,000 equipment facility secured by the ovens, with each drawdown tied to a milestone.
- Set a covenant that cash never falls below one month of payroll, tested monthly against the closed management accounts rather than the bookkeeper's draft figures.
- Ask the owners for a written grocer renewal or a shorter contract extension before the first tranche is released, and record the condition in the approval memo.
Results
The committee approved a 600,000 shop loan with the first tranche held until the grocer's renewal was signed. The equipment facility of 250,000 was approved on a separate memo. Two months later the reconciled statements showed that late supplier postings had been moving roughly 30,000 of expenses each month into the following period. The corrected margins were lower than the owners had presented, but still within the covenant. The first tranche was released after the grocer signed a three-year renewal, and the cash covenant has been tested each month since.
Questions for discussion
- Which downside assumption would you stress first, and what evidence would move you from a range to a point estimate?
- How should a lender weigh a borrower's projected break-even against a record of closed months that does not yet exist?
- What does a split facility, with equipment financed against resale value, change about the risk the committee is actually taking?
Case Study 5
Ranking Control Findings Across a Clinic Group
Marlow Sound Clinics Group, a network of seven outpatient clinics with a shared finance office, about 420 staff, and a central accounts payable team.
The situation
The internal review covered the last close cycle across all seven clinics and the shared finance office. It produced nine findings, and five of them mattered for the year-end numbers. Refunds to patients totaling 186,000 were released without a second approver, which the policy required above 500. Accrual cutoffs were missed at two clinics, leaving liabilities understated by 41,000. The vendor master carried 17 duplicate records, and one clinic's bank reconciliation had not been signed off for three months. Expense claims under 1,000 were approved by the submitter in a handful of cases, totaling 9,600 for the year. Nkechi, the head of internal controls, had to rank the findings and name owners before the audit committee met.
The challenge
The findings did not carry equal weight. The refund gap was large but had happened only in a few batches. The cutoff error was smaller in dollars but touched every period end. Owners were scattered across clinic managers, the central payables lead, and the bookkeeper for one site, and several had already started fixing things on their own, which made it hard to tell what closure would look like.
Analysis
Nkechi ranked the findings on two axes: the size of the effect on reported results, and the likelihood that the same weakness would recur in the next cycle. Materiality was judged against the group's planned threshold for the year-end statements, not against whatever amount felt alarming. The refund approvals ranked highest on effect because 186,000 was large relative to the period's operating result. The cutoff misses ranked high on recurrence because they sat in a process every clinic runs each month. The duplicate vendors were low on effect but worth tracking, since each one is an opening for a misdirected payment. Transaction records were pulled for each finding to confirm the amounts before any ranking was published.
Owners were assigned by process rather than by location. The refund approval gap belonged to the central payables lead, Oskar, because the approval path ran through that team's system. The cutoff procedure belonged to the group controller, with each clinic manager signing a close checklist. Closure was defined as evidence, not as a statement that the fix was in place. The refund finding would close only after three consecutive month-end batches showed two approvers on every payment above the threshold, sampled from the system log. The cutoff finding would close when accruals at all seven clinics matched subsequent invoices by the close deadline, tested by a reviewer outside the preparing team.
Actions taken
- Review each refund batch against patient records and recover any overpayment, then configure the payables system to block release above 500 without a second approver.
- Assign the cutoff procedure to the group controller, with an accrual listing per clinic reviewed by someone outside that clinic before the close deadline each month.
- Merge the seventeen duplicate vendor records after confirming bank details for each, and restrict vendor creation to a team that cannot also approve payments.
- Set closure tests for each finding: three clean month-end samples for refunds, a full accrual match for cutoffs, and a signed reconciliation sample for every clinic bank account.
Results
Six of the nine findings closed within two cycles. The refund batches were reviewed and 3,200 was recovered from overpayments, and the system block was tested by attempting a single-approver release, which failed as designed. The accrual listing caught a 12,000 liability at one clinic in the first month. The vendor merge reduced duplicates to zero. The cutoff finding stays open until a third clean month is sampled, and the audit committee accepted that as the closure test.
Questions for discussion
- How would you change the ranking if a low-recurrence finding had a much larger effect than the sum of several smaller recurring ones?
- Who should own a control that spans three teams, and what happens to accountability if that owner also performs the test?
- What evidence would convince you that a finding is closed, rather than simply no longer appearing in the review?
Learning Resources
Finance CASE STUDIES
China, India, United States, Indonesia, Pakistan, Nigeria, Brazil, Bangladesh, Russia, Japan, Mexico, Philippines, Vietnam, Egypt, Turkey, Germany, United Kingdom, France, Italy, Canada, South Korea, Australia, Spain.