Creative Business, Portfolio and Client Management Case Studies
Creative5 Case StudiesUpdated Sep 2026
Creative businesses rarely fail because the work is poor. They fail when the money behind the work drifts out of view: invoices that do not match the ledger, forecasts built on hope, a deposit from an account nobody expected, or a review that finds the same gap twice. The five cases here come from studios, collectives, and agencies that manage portfolios of client projects. Each begins with a records problem, then asks what exposure it created, what evidence the team could rely on, and where the decision thresholds should have sat.
Case Study 1
Quarterly receivables gap traced to manual journals
Quillon Brand Works, a nine-person brand and packaging studio that bills retainers monthly and pays freelance illustrators and print vendors through one shared ledger.
The situation
Quillon Brand Works is a nine-person brand and packaging studio that bills retainers monthly and pays freelance illustrators, photographers, and print vendors through a shared ledger. In the second month of the quarter, Ines, the studio's operations lead, reconciled the billing subledger against the general ledger and found a gap of 14,450 in client receivables. The billing export showed 412,600 invoiced, while the ledger recorded 398,150. Nobody had flagged it because each report looked internally consistent. Two retainers had been partly credited in the ledger but not in the billing system, and the bookkeeper, Tomasz, had approved the monthly close without opening the subledger detail.
The challenge
The variance was small against quarterly revenue, but it sat inside receivables, which the owner used to decide how much could go to freelance commissions. Ines needed to know whether the gap was timing, a keying error, a credit note posted in one system only, or a duplicate invoice. The supporting documents were spread across invoice files, client remittance emails, and a payment processor export that used different customer codes. She had until the next commission approval to give a clear answer.
Analysis
Start with what each record claims. The billing subledger says a document was issued to a client. The general ledger says revenue was recognized and a receivable was booked. The processor export says money moved. Those are three different assertions, and a gap between them is not automatically an error. A timing difference and a duplicate invoice look identical in a single total, so each open item has to be mapped to its source before any number is changed. The exposure here was modest in cash terms, but the commission budget depended on receivables being right, so the risk lay in the decision the number fed.
The trace worked backward from the ledger. Ines matched each of the 61 open invoices against remittance records and found that nine had been credited through a manual journal rather than through the billing system, which explained most of the gap. Two others were duplicates created when a retainer was reissued after a change of client contact. The evidence was strong enough to reverse the journal and void the duplicates, but she set a rule for herself: any unexplained residual above 1,000 would stop the commission approval until the owner signed off. The residual came to 2,300, so the approval waited.
- Extract subledger Pull billing documents and customer codes at period close.
- Match to ledger Trace each open item to a posting, credit note, or receipt.
- Clear or escalate Reverse proven errors; route residuals above threshold to the owner.
- Sign and file Store the signed worksheet with the period label for review.
Actions taken
- Require every manual journal to cite a billing document number and a second approver before posting. Ines drafted the rule and the bookkeeper tested it on the next close.
- Build a shared customer identifier across billing, ledger, and processor exports so that a payment can be traced to one client without manual matching. Tomasz mapped the 47 active client codes first.
- Set a written threshold: any unexplained residual above 1,000 in receivables blocks discretionary spending decisions until the owner reviews the reconciliation.
- Run the subledger to ledger tie-out in the first week after each close, not at quarter end, and keep the signed worksheet in the shared drive with the period label.
Results
The reversal cleared 12,150 of the 14,450 gap, leaving a residual of 2,300 that the team documented as a remittance paid in two parts and not yet matched. The owner approved a smaller commission budget for the following month and held the rest until the remittance was traced. The tie-out took under a day in the next close, and no manual journal was posted without a billing reference. The studio found no fraud, but the review showed that the close had relied on each report looking reasonable rather than on documents agreeing with one another.
Questions for discussion
- Which of the three records in this case should be treated as the authority for receivables, and what would change that choice?
- How would you set a residual threshold that is low enough to catch real problems but high enough that the close does not stall every month?
- What evidence would convince you that a manual journal was a legitimate correction rather than a way to hide a timing problem?
Case Study 2
Forecast tested after two anchor clients paused work
Saltmarsh Motion Lab, a fourteen-person motion design studio that sells multi-month production slates to streaming and brand clients.
The situation
Saltmarsh Motion Lab is a fourteen-person motion design studio that sells production slates, usually three to five projects running at once, to streaming and brand clients. Its plan assumed that two anchor accounts would keep their scheduled deliveries through the year. In the sixth week of the quarter, one anchor client paused an 18-episode series pending an internal budget review, and the second cut its order by a third. Monthly fixed costs stood at 86,400, including payroll for eleven staff and the lease on the stage. Bridget, the managing director, had to decide how to respond before the next payroll run.
The challenge
The forecast showed quarterly revenue falling from a planned 540,000 to near 390,000, which would leave cash covering about five weeks of fixed costs without new work. The team did not know how firm the remaining orders were, and the pipeline held proposals at different stages with different odds. Bridget needed scenarios she could defend to the bank, to the staff, and to herself, along with a clear view of which assumptions the whole picture depended on.
Analysis
The first task was separating commitments from expectations. Signed statements of work, deposits received, and milestone schedules went in one column. Proposals, verbal interest, and renewals still under discussion went in another. The original plan had counted both as revenue, which is common in creative work where one strong relationship can make a quarter look secure. Once the records were split, the contracted base sat at about 61 percent of plan, a clearer picture than any single probability could give. Teams grounded in Creative Business, Portfolio and Client Management usually rebuild the forecast from documents first and add judgment afterward, so every assumption stays visible when someone challenges it.
Three scenarios were built on the contracted base. In the base case the paused client returned next quarter and the second held a reduced order, putting revenue near 430,000. The downside assumed the pause ran all quarter and two proposals fell away, bringing revenue near 360,000. The severe case kept only the smallest retainer. The assumption that mattered most was how long payroll could run without billable work, so the team priced bench time, moved two designers to internal brand work, and required deposits on new proposals. The trigger was agreed in advance: contracted revenue below 60 percent of fixed costs over the next six weeks would reduce hours.
Actions taken
- Rebuild the forecast from signed documents first, then add pipeline work at stated probabilities that the sales lead must defend in writing.
- Ask the paused client for a revised schedule in writing, and tie any staffing change to the date that schedule confirms.
- Move two designers onto internal brand work for six weeks and price any remaining bench time, so the payroll decision rests on a known cost.
- Set a written trigger: if contracted revenue for the next six weeks covers less than 60 percent of fixed costs, reduce hours and freeze discretionary spending.
Results
Bridget took the base case to the bank with the scenario table, and the lender extended the existing line for another quarter without new collateral. The paused client resumed a smaller series the following quarter, and the second held its reduced order. Revenue landed near the base case and the trigger was never reached, although it came close in the fifth week. The staff kept their schedules, and the studio stopped counting proposals as revenue until a deposit arrived.
Questions for discussion
- Which single assumption in this forecast would you test first, and what evidence would change your confidence in it?
- How should a studio decide between reducing hours now and waiting for a client decision that has no firm date?
- What would have to be true in the pipeline before you counted a proposal as revenue in the base case?
Case Study 3
Third-party deposit and an unexpected refund request
Linden Row Photography Collective, a cooperative of six photographers and two producers that books commercial shoots and holds client deposits in a shared account.
The situation
Linden Row Photography Collective is a cooperative of six photographers and two producers that books commercial shoots and holds client deposits in a shared account. A new client, a consumer goods company registered abroad, asked to book a three-day shoot with a deposit of 48,000, paid by transfer from a third party's account in another country. Two days later the client requested a refund of 41,000 to a different bank, citing a change in campaign timing. Hallam, the collective's treasurer, noticed that the payer name did not match the contract and held the refund.
The challenge
The collective had no written policy for third-party payments or refunds to new accounts, and the producer who onboarded the client had already scheduled crew. Hallam needed to decide whether the pattern was a legitimate change of plans with an unusual payer, or a warning sign that the shared account might be used to move money. Holding the refund risked annoying a client, and releasing it risked the bank's confidence in the account. The crew had turned down other work, and the decision had to be made within the week.
Analysis
The indicators were weak one by one but formed a pattern. A payer that differs from the contracting party, a deposit followed closely by a refund request, and a payout to an account not on file each have ordinary explanations. Together they raise a harder question: whether funds are passing through the collective rather than being earned by it. The exposure is not only the 41,000 refund. It is the collective's standing with the bank that holds the shared account and monitors unusual movement. A careful review starts from what the documents show, not from what the client says it intends.
Documentation was the weakest point. The contract named the client, but the signed order carried no company registration number, the billing address matched a shared office, and the invoice had come from a free webmail address. The collective's only control on deposits was a single signature, with no rule for third-party funds. Hallam set escalation thresholds in advance: any refund to an account not in the original payer's name, or any third-party payment above 5,000, would stop the transfer and go to the cooperative's outside accountant and the bank's compliance contact. The refund stayed on hold, and the shoot moved to a conditional booking rather than a cancellation.
Actions taken
- Hold the refund and request in writing the payer's identity, the reason for the third-party payment, and a bank confirmation of the refund account.
- Obtain company registration details and a contract signed by an officer, then check them against public registers the client cannot edit.
- Adopt a written policy that routes any third-party payment or change of refund account to the treasurer and outside accountant before funds move.
- Report the pattern to the bank's compliance contact through the agreed channel if the client cannot document the payer and refund account within the review window.
Results
The client did not supply an officer-signed contract and did not explain the payer, so the collective kept the 48,000 deposit on hold and offered to reschedule the shoot once documents arrived. No further contact came. The collective filed the pattern with its bank through the agreed channel, and the bank asked for three months of transaction history, which the treasurer supplied within the week. The crew was released to other work. The collective now runs a checklist on every deposit above 5,000 and treats refund requests as a separate approval.
Questions for discussion
- At what point would you stop weighing a client's explanation and treat the documentary gap alone as the decision point?
- How should a small collective balance its duty to a bank against the commercial cost of losing a client?
- Which control would you add first to a shared account used by several independent freelancers, and why that one?
Case Study 4
Working capital request from a textile design house
Lantern Hill Textiles, a twenty-two-person design house that prints patterns for fashion labels and home goods brands.
The situation
Lantern Hill Textiles is a twenty-two-person design house that prints patterns for fashion labels and home goods brands. It has grown through commissioned collections and now holds a licensing offer from a homewares retailer that would place its patterns across 140 stores. The offer requires an upfront minimum order and a 90-day payment term. The studio asked the Brightwater Credit Cooperative for a 220,000 working capital facility, secured against inventory and the licensing receivable. Pilar, the credit officer, received the application with three years of management accounts and a draft license agreement.
The challenge
The founders presented the license as transformative, and their forecast showed margins improving with volume. Pilar had to decide whether the facility suited a business whose income had been lumpy, whose largest previous client had paid late, and whose inventory was valued on the founders' own estimates. The committee would not meet again for three weeks, and the retailer's order window would close before then, so waiting was itself a decision.
Analysis
The management accounts showed revenue rising steadily, but gross margin had moved by several points between quarters, and two of the last four quarters closed with cash below one month of payroll. The forecast assumed full store coverage from the first quarter of the license, which is rarely how rollouts go. Pilar rebuilt the case from bank statements instead of the management pack, and checked aging against the largest client's remittance history. The downside case assumed a staged rollout to half the stores, a 120-day payment term, and a 15 percent markdown on inventory held for the retailer.
Suitability depended on whether the facility matched the cash cycle it was meant to bridge. A line sized to the minimum order would cover production but leave the 90-day gap open if the retailer paid late. Pilar tested three structures. A single lump sum did not fit, and a revolving line ran past the payment window. A smaller tranche tied to the signed purchase order, with a second tranche released on first accepted delivery, did. The conditions she recommended were a signed purchase order, an independent inventory count, personal guarantees from both founders, and a covenant keeping drawn balances below 60 percent of confirmed receivables.
- 1Application receivedManagement accounts, bank statements, and draft license reviewed together.
- 2Downside testingStress tests on rollout pace, payment terms, and markdown exposure.
- 3Structure fitTranches matched to the purchase order and delivery cash cycle.
- 4Conditions setSigned order, inventory count, guarantees, and a receivables covenant.
- 5Release or declineTranche two waits for accepted delivery; thresholds decide the outcome.
Actions taken
- Require a signed purchase order from the retailer before the first tranche is released, and confirm the order volume with the buyer directly rather than through the applicant.
- Commission an independent inventory count and value stock at cost less expected markdown, not at the founders' estimate of retail value.
- Release the facility in two tranches, the second tied to first delivery and accepted payment, so drawn balances follow the cash cycle.
- Set a decline threshold: if confirmed orders cover less than 70 percent of the requested amount, or payment history shows settlement beyond 60 days, decline.
Results
The committee approved a reduced facility of 160,000 in two tranches, with the second conditional on the first delivery being accepted and paid within the stated window. The retailer confirmed an order covering 74 percent of the requested amount, just above the threshold. The first tranche funded production, and the first delivery was paid in 49 days. The second tranche followed a month later. Lantern Hill has since asked for a review of covenant levels, which the cooperative will consider when the license reaches full rollout.
Questions for discussion
- Which of the three downside assumptions would you stress further, and what evidence would you ask for on it?
- When is a tranched facility the better answer, and when does it make the borrower's planning harder?
- How much weight should a lender give personal guarantees when the main asset is a body of design work?
Case Study 5
Billing control findings from a multi-office agency review
Fennick Studio Group, a design agency with 48 staff across three offices that bills project work from time sheets, purchase orders, and a client portal.
The situation
Fennick Studio Group is a design agency with 48 staff across three offices, billing project work from time sheets, purchase orders, and a client portal. After a client disputed an invoice, the board asked for an internal review. Anouk, the reviewer, sampled 40 projects closed over the last two quarters. She found that time entries were not approved before invoicing on roughly a third of sampled projects, and several write-offs had been approved by the same person who raised them. Anouk had five working weeks to report before the next board meeting.
The challenge
The findings were spread across offices and systems, and no single manager held the full picture. Some issues plainly touched revenue, such as unapproved hours billed at full rates. Others were design gaps that had not yet produced a loss. The board wanted a ranked list, named owners, and tests that showed the fixes worked, not just that policies had changed. Management had a habit of closing findings by updating a document.
Analysis
Ranking began with exposure, not with how alarming each finding sounded. Unapproved time billed at full rates touched the revenue line directly and appeared on a large share of sampled invoices, so it ranked first. The single-approver write-offs ranked second because they removed value from receivables without a second check, even though the sampled amounts were small. Purchase orders raised after work started ranked lower on cash value but higher on the chance that a supplier charge was never authorized. Materiality here was a judgment about what the board needed to know, weighed against how far each gap could move the reported result.
A finding closes when a test shows the control operates, not when a policy changes. For time approval, the test drew a fresh sample of invoices issued after the change and confirmed that each billed line carried an approver's name and a timestamp earlier than the invoice date. For write-offs, the test required two different people in the approval log for every item above 2,500. Anouk asked for the test to be repeated in a later period, so management could not reach closure on a single clean sample. Each test set a pass rate target and a rule for what counted as failure.
- In the first weekFreeze billing on projects with unapproved time; name owners for each finding.
- By month endSecond-approver rule live; portal credentials reissued to named users.
- In the second monthFirst sample test drawn; pass rate compared against the target.
- In a later periodRepeat test run without warning; any failure reopens the finding.
- Before the board meetingClosure file tabled with samples, sign-offs, and remaining exceptions.
Actions taken
- Block invoice generation for any project with unapproved time entries, and assign the billing lead as owner with a target of closing the gap in the first month.
- Require a second approver for write-offs above 2,500, with the operations director as owner and a monthly exception report to the finance committee.
- Move supplier purchase orders ahead of work orders in the portal, owned by the procurement lead, and reject invoices that arrive without a matching order number.
- Retire the shared portal login, issue named credentials to each account manager, owned by Yusuf, the systems administrator, and test access quarterly against the staff roster.
Results
Management reported the billing gap closed in the second month after the fix, and the first sample test passed at 96 percent. The second test, drawn in a later period, found three invoices still carrying unapproved hours, which reopened the finding and led to a credit note. The write-off rule caught two items that would previously have passed. Named portal credentials went to all 19 account managers, and each closure file now holds the test sample, the reviewer's sign-off, and the retest result.
Questions for discussion
- What would you need to see before accepting a control closure when the original failure was sporadic rather than systematic?
- How should a review decide between fixing a finding in one office and redesigning the control across all three?
- Which of the five findings would you rank differently, and what exposure would justify the change?
Learning Resources
Creative 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.