A deal that went sideways: what we learned

At 6:12 on a Tuesday morning, the property manager opens a spreadsheet with 23 vacant units, 11 unfinished turns, and $41,700 of invoices due Friday.

The investor update still says the plan is “progressing.” The bank account says the plan has about six weeks left.

This is an explicitly labeled composite hypothetical, assembled from common multifamily failure patterns for education. It is not a PR deal, no actual property or investor result is being described, and the numbers are invented. The point is not to gossip over a wreck. The point is to find the exits everyone drove past.

The manager calls the operator.

“Do you want the vacancy number or the cash number first?”

That is the moment the deal stops being a renovation story.

The opening scene looked ordinary

Picture 160 apartments purchased for $17.92 million, or $112,000 per unit. The acquisition model assumes $13.44 million of interest-only, floating-rate debt, 75% loan-to-value, priced at SOFR plus 3.25%. Equity funds the down payment, closing costs, $600,000 of reserves, and an $8,000-per-unit renovation program.

The opening script is clean: renovate eight units a month, raise average rent from $1,100 to $1,275, hold 95% economic occupancy, and sell after stabilization.

The source file is not the glossy model. Before believing those inputs, inspect the purchase and sale agreement and amendments, dated rent roll, every lease and concession addendum, tenant-ledger aging, security-deposit ledger, trailing-12 and trailing-3 operating statements, general ledger, 12 months of bank statements, property-tax bills, insurance loss runs and binding quote, payroll register, utility bills, vendor contracts, and delinquent accounts-payable report.

The current Fannie Mae rules for Underwritten Net Cash Flow call for objective income and expense measures, historical operating statements, recent vacancy history, and separate treatment of vacancy, concessions, and bad debt. That is lender guidance, not a magic acquisition formula. The discipline is useful because signed rent is a promise; the bank deposit is the receipt.

Five assumptions became one problem

At acquisition, the model stressed one variable at a time. Slower rent growth was tested on Monday, higher expenses on Tuesday, and a later sale on Wednesday. The risks never had to share a week.

The property scheduled them all for Friday.

Here, five things move together:

1. Renovations average five units a month, not eight.
2. Each completed turn costs $12,500, not $8,000.
3. Economic occupancy falls while units sit offline and concessions rise.
4. Insurance, repairs, payroll, and utilities run above budget.
5. Floating debt resets while property income is weakening.

The New York Fed publishes the official SOFR series and methodology. The index is not a footnote when the loan floats. It gets a monthly vote on the checking account.

Friday’s cash meeting

The stabilized acquisition case begins with $2,166,400 of effective gross income and $1,280,000 of operating expenses, leaving $886,400 of NOI. At a 4.00% all-in interest rate, interest-only debt service is:

Acquisition case	Arithmetic	Amount
NOI	$2,166,400 - $1,280,000	$886,400
Annual debt service	$13,440,000 x 4.00%	$537,600
Cash before capital work	$886,400 - $537,600	$348,800

Now run the combined miss. NOI drops to $720,000. The all-in rate reaches a hypothetical 7.50%, so debt service becomes $13,440,000 x 7.50% = $1,008,000. Operations burn $288,000: $720,000 - $1,008,000.

Forty completed turns cost an extra $4,500 each, another $180,000. A required roof and drainage scope consumes $100,000. Total reserve use is $288,000 + $180,000 + $100,000 = $568,000. The original $600,000 reserve is now $32,000 before the next surprise.

At the meeting, somebody calls it a temporary timing issue.

The controller turns the laptop around. The reserve balance is $32,000 before the next surprise.

The controller has ended the timing debate. Timing problems own calendars. This problem owns the capital structure.

The first warning was in the work file

Capex did not come from nowhere. The property-condition assessment, roof inspection, sewer scope, unit walk sheets, environmental report, contractor bid tabs, permits, change orders, invoices, lien waivers, and draw requests contained the clues. Fannie Mae’s current Property Evaluation update revised its property-evaluation requirements and Instructions for Performing a Multifamily Property Condition Assessment, Form 4099. That is a useful level of specificity: name the inspection, the repair, the cost, and the document that controls it.

The operator also had weekly vacancy reports, make-ready aging, work-order backlog, leasing traffic, application denials, renewal offers, trade-out reports, collections, payroll variance, and budget-to-actual financials.

“We are behind schedule” would have been useful in week two. By week twelve, it was merely polite. A renovation schedule only becomes an operating plan when someone reconciles it to labor capacity, occupancy, and cash every week.

Cash is low; authority matters now

When cash gets tight, read the promissory note, loan agreement, mortgage or deed of trust, guaranty, interest-rate-cap agreement and confirmation, cash-management agreement, reserve agreement, lender covenant calculations, and every lender notice. Then read the PPM, operating agreement, subscription agreement, sources and uses, distribution waterfall, capital-call provisions, sponsor-loan authority, removal rights, and reporting requirements.

The OCC Commercial Real Estate Lending handbook identifies interest-rate, liquidity, operational, and credit risk as connected parts of CRE lending. Investors should make the same connection. Counsel, lenders, accountants, and qualified property professionals must interpret the actual documents and facts; this lesson does not do that for them.

Once cash is tight, the question “What should we do?” is incomplete. Add: “Who is allowed to do it, under which document, before what date?”

The update loses its adjectives

The failure mode is not merely missing the original plan. It is reporting adjectives after the numbers changed.

In the next draft, “progressing” gets deleted. So does “temporary pressure.” The update now shows current cash, monthly burn, unpaid bills, cap status, covenant position, revised completion dates, and management’s recommended decision. The tone gets worse. The report gets better.

Hard questions belong in the room early:

• What is unrestricted cash today, and how many weeks does it cover at the current burn?
• Which rent premiums are supported by signed leases and net collections after concessions?
• What does debt service become at each cap strike, extension, or expiration date?
• Which capital projects are safety-critical, lender-required, or optional?
• What covenant is closest to failure, and when must the lender be notified?
• Who can authorize a capital call, sponsor loan, renovation pause, sale, or loan workout?
• What fact would make management abandon the original business plan now?

The plan gets thirteen weeks

The concrete next move is a 13-week cash forecast built from the bank balance, not the acquisition model. List rent collections, payroll, utilities, insurance, taxes, debt service, capex commitments, payables, lender reserves, and minimum operating cash by week. Tie every starting number to a bank reconciliation, invoice, contract, or ledger. Update it every Friday.

Then choose. Pause nonessential turns, protect occupied units, negotiate with vendors and the lender, inject capital under the governing documents, or sell while choices still exist. None is painless.

The transferable rule is to combine the misses before the reserve starts paying for them. Run renovation pace, cost overruns, occupancy loss, expense growth, and floating debt in the same downside case. If the answer is $32,000 before the next surprise, the decision date is already behind you.