SPA in mergers and acquisitions: lessons from a broken deal

The deal is the handshake, the spreadsheet, the management presentation, and the promise of “strategic alignment.” The SPA is where those promises are translated into conditions, covenants, representations, remedies, and closing mechanics. It is also where the weaknesses in the original story become impossible to ignore.

Every comma in that document is usually a scar from a previous transaction that went wrong. A broad definition became an argument. A vague covenant became a consent dispute. A working capital target became a price fight. If you do not read the agreement as a map of those risks, you are not protecting the transaction. You are merely documenting the hope that it will close.

A frequently cited market statistic puts the failure rate for signed letters of intent at roughly one-third. That figure describes the proportion of LOIs that do not ultimately reach closing; it should not be presented as a calculation from Axial’s separate sample of 75 transactions that collapsed in 2025.

The 2026 Axial Dead Deal Report is useful for a different reason. Its sample consists of failed transactions, and its category breakdown shows where those failures were attributed. It does not tell us how many transactions in the broader market closed, and it would be wrong to infer a closing majority from a sample made up of dead deals. The more precise lesson is this: a signed LOI creates momentum, not certainty. The transaction still has to survive diligence, financing, negotiation of definitive documents, interim operations, and the closing conditions in the SPA.

The deals that fail rarely do so because of one dramatic sentence in the agreement. More often, the SPA exposes a problem that was already present in the numbers, the operating model, or the parties’ assumptions about control.

The Anatomy of Deal Failure: Why 33% of LOIs Never Close

The headline number obscures the more useful question: why do transactions die after the parties have already agreed on a framework and, often, a headline price?

The reported one-third LOI failure figure should be kept analytically separate from Axial’s sample of 75 failed 2025 transactions. Within that failed-deal sample, 46.6% of the reported causes were attributed to diligence findings when non-QoE diligence findings and QoE EBITDA discrepancies are considered together. The category breakdown was:

• Non-QoE diligence findings: 25.3%
• QoE EBITDA discrepancies: 21.3%
• Renegotiation challenges: 14.7%
• Seller decisions: 13.3%
• Financing constraints: 10.7%
• Business underperformance: 8.0%

“Roughly half of the reported dead deals died in diligence. The rest still had to survive financing, negotiation, performance, and the parties’ willingness to keep going.”

That distinction matters because an LOI is generally a preliminary document. It may record the parties’ commercial understanding and establish an exclusivity period, confidentiality obligations, access rights, or other provisions that are expressly binding. But the headline purchase price and the proposed transaction structure are commonly subject to negotiation of definitive documents and completion of diligence. Signing an LOI does not, by itself, transfer ownership or put the seller under a buyer’s general operating-control regime.

The practical risk is not that the LOI has already sold the company. The risk is that both sides begin behaving as though the deal is inevitable. The seller invests in the proposed transaction, gives the buyer access to sensitive information, and starts making decisions with the closing date in mind. The buyer starts allocating internal resources, discussing financing, and treating the target’s projections as the base case. By the time the SPA is being negotiated, a material disagreement can feel like a betrayal rather than what it actually is: a risk that was not resolved before the parties moved too far forward.

A failed LOI can reflect several different kinds of failure:

1. The business was not what the buyer understood it to be. Customer concentration, churn, regulatory exposure, deferred maintenance, litigation, or weak internal controls can change the risk profile without changing the headline story.

2. The earnings number was not sufficiently durable. An add-back may be defensible in a management presentation and still fail the buyer’s quality-of-earnings methodology.

3. The parties agreed on price but not on price mechanics. A headline figure can conceal disagreement over debt, cash, working capital, earnouts, rollover equity, or the treatment of transaction expenses.

4. The seller and buyer never agreed on the level of interim control. The seller may expect to run the business as before; the buyer may expect consent over every material decision.

5. The business deteriorated during the process. A transaction can be attractive at signing and less attractive by closing if performance, financing conditions, or a key customer relationship changes.

The fact that a signed LOI exists should not encourage complacency. It should sharpen the work done between the LOI and the SPA. That is the period in which the parties convert a broad commercial understanding into an enforceable allocation of risk.

Financial Diligence and the Rising Threat of QoE Discrepancies

Let me translate the financial issue into plain English, because the consulting decks and law-firm memoranda often bury it under terminology.

When a buyer signs an LOI, the proposed price is typically based on a financial picture supplied by the seller and discussed during the initial process. The picture may be well prepared and honestly presented. It may also contain adjustments that require further testing. Discretionary expenses may be added back. Owner compensation may be normalized. One-time revenue may be treated as non-recurring. A cost that appears unusual in isolation may turn out to be a normal cost of operating the business.

That is what the quality-of-earnings process is designed to examine. A QoE provider is not simply checking whether the arithmetic adds up. It is testing whether the reported earnings measure reflects recurring economic performance and whether the assumptions used to build adjusted EBITDA can survive scrutiny.

A useful QoE review asks questions that sit between accounting and operations:

• Is the claimed adjustment genuinely non-recurring, or has it appeared in several periods under different descriptions?
• Does the business need to incur the expense again after closing, even if the seller did not incur it in the same form?
• Are revenue-recognition practices consistent across customers and reporting periods?
• Are customer credits, rebates, returns, or implementation costs being recorded in the right period?
• Does the margin profile depend on a small number of customers, contracts, employees, or suppliers?
• Are recent results supported by cash collection, or only by reported revenue?
• Is the working capital balance consistent with the level required to operate the business after closing?

The Axial data identified QoE EBITDA discrepancies as 21.3% of the reported causes of failed 2025 transactions, compared with 10.6% in 2023. That comparison is useful as a warning about the category, but it does not establish why the share changed or prove a long-term market trend. It does show how quickly a disagreement over adjusted EBITDA can move from a diligence issue into a transaction issue.

The second row deserves particular attention because adjusted EBITDA is not merely a presentation metric. It can influence enterprise value, debt capacity, leverage ratios, earnout thresholds, management incentives, and the buyer’s return model. A disagreement over one add-back can therefore spread through the entire transaction.

The SPA cannot repair a financial model that the parties never agreed on. It can, however, determine what happens when the model proves incomplete. The agreement may include detailed representations about financial statements, revenue, receivables, customers, inventory, liabilities, and the absence of undisclosed obligations. It may also allocate risk through indemnification, purchase-price adjustments, escrows, or specific remedies. Those provisions are not substitutes for diligence. They are the legal consequences of what diligence did or did not uncover.

From EBITDA disagreement to purchase-price renegotiation

The most difficult moment often arrives after the buyer has invested heavily in the process but before the definitive SPA is signed. The buyer may say that the QoE report changes the earnings base and therefore the price. The seller may say that the buyer is changing the rules after receiving the benefit of a competitive process.

Neither position can be evaluated in the abstract. The parties need to identify the exact source of the difference:

• Is the issue a factual error in the seller’s financial information?
• Is it a difference in accounting policy?
• Is it a disagreement about whether an expense is recurring?
• Is it a question of whether the buyer will operate the business differently after closing?
• Is the buyer applying a multiple to a revised EBITDA figure that the seller never accepted?

The negotiation becomes more productive when the parties separate these questions. A seller may be able to defend an adjustment while conceding that the buyer’s underwriting should be more conservative. A buyer may accept the historical accounting while refusing to value a customer relationship as recurring. The SPA then becomes the place to document the agreed treatment rather than a device for hiding an unresolved valuation dispute.

A common mistake in M&A SPA negotiation is to focus on the purchase-price number and postpone the mechanics. That is how a transaction can appear agreed while the parties remain divided on what the number includes. “Cash-free, debt-free” does not answer every question. The parties still need to address transaction expenses, unpaid bonuses, deferred revenue, tax liabilities, leases, customer deposits, normalized inventory, and the level of working capital required at closing.

“Adjusted EBITDA is not a verdict. It is an argument about what the business will earn after the buyer takes control.”

The “Ordinary Course” Trap: Lessons from AB Stable and Operational Control

Nowhere does the SPA in mergers and acquisitions show its teeth more clearly than in the ordinary-course covenant.

The covenant usually governs the period between signing and closing. Its basic purpose is straightforward: the seller remains in control of the business during that period, but agrees to operate it in a way that is consistent with its past practice and the agreed transaction assumptions. The details are not straightforward. They determine how much freedom the seller retains and which decisions require the buyer’s consent.

This is where the distinction between an LOI and an SPA becomes legally important. Signing an LOI does not automatically subject the seller to the definitive agreement’s interim operating covenants. Those obligations generally arise when the parties sign the SPA or merger agreement, and only to the extent the agreement makes them binding. An LOI may contain its own conduct-related provisions, but they cannot simply be assumed.

Once the SPA is signed, however, the seller’s room to maneuver can narrow quickly. The covenant may address:

• changes to compensation, benefits, or headcount;
• material contracts and customer arrangements;
• pricing, discounts, rebates, and credit terms;
• capital expenditures and acquisitions;
• borrowing, guarantees, and liens;
• litigation, settlements, and regulatory commitments;
• accounting policies and changes to working-capital practices;
• business combinations, divestitures, or new lines of business.

The Delaware Supreme Court’s decision in AB Stable VIII LLC v. MAPS Hotels and Resorts One LLC remains an important warning. The case involved the sale of a hotel business during the COVID-19 pandemic. The court held that the seller’s extensive operational changes did not satisfy the ordinary-course covenant because the business had not been operated in the ordinary course consistent with past practice, even though the seller argued that the changes were commercially necessary in extraordinary circumstances.

The case does not mean that a seller must ignore a crisis or refuse to protect the business. It means that necessity is not automatically a contractual defense. The seller must understand what the covenant says, what exceptions it contains, and whether buyer consent is required. If the business is facing an event that makes historical operations impossible, the safest course is usually a documented consent process, not an assumption that a court will later sympathize with the decision.

Operational control is a drafting problem

The ordinary-course covenant can be drafted at a high level or with considerable precision. A seller will typically seek flexibility to respond to changing market conditions. A buyer will seek visibility and consent rights over decisions that could affect value before closing. The negotiation should distinguish between decisions that genuinely change the business and decisions that are routine consequences of operating it.

For example, a blanket requirement to obtain consent for every material customer discount may give the buyer too much practical control. A covenant that permits all pricing changes without limitation may leave the buyer exposed to an artificial boost in short-term revenue. The right answer depends on the company’s business model, sales cycle, seasonality, and history of pricing decisions.

The same is true of capital expenditure. A fixed dollar threshold may be useful, but it should be tested against the target’s normal spending patterns. A threshold that looks large in a term sheet may be immaterial for a manufacturing business and enormous for a software company. The drafting should also address emergency expenditures, repairs, regulatory compliance, and projects already approved before signing.

The consent process matters as much as the list of restricted actions. The SPA should make clear:

1. Who may request consent on behalf of the seller.

2. Who may grant it on behalf of the buyer.

3. Whether consent may be withheld in the buyer’s sole discretion.

4. How quickly the buyer must respond.

5. Whether silence counts as consent.

6. How an emergency decision is documented.

7. Whether the buyer’s consent creates liability or merely waives a covenant objection.

A buyer that insists on broad consent rights but responds slowly can create operational paralysis. A seller that treats consent as a formality can create a breach record before closing. Neither outcome is good deal management.

The covenant does not care about a founder’s intuition or a banker’s assurance that “everyone understood the arrangement.” It cares about the words the parties signed and the evidence showing how they applied them.

Testing the Limits of Material Adverse Effect Clauses in Court

Buyers, naturally, would prefer a broader escape hatch. The Material Adverse Effect clause, or MAE, is intended to address a serious deterioration in the target or its prospects between signing and closing. It is often described as the agreement’s nuclear option. That description is useful only if it does not encourage buyers to treat the clause as a general right to walk away from a disappointing transaction.

MAE provisions are heavily negotiated because they sit at the boundary between transaction risk and ordinary business risk. A buyer wants protection against a fundamental change in the target. A seller does not want to insure the buyer against every adverse quarter, interest-rate movement, supply-chain problem, or industry-wide shock.

Courts have generally approached MAE claims cautiously. The question is not whether the business performed below the buyer’s expectations. The question is whether the contract, properly interpreted, allocated the relevant risk to the seller and whether the alleged change is sufficiently material and durable to satisfy the clause.

That makes the drafting architecture critical. A typical MAE definition may contain:

• a general statement describing a material adverse effect on the business, assets, results, or prospects;
• exclusions for broad market, industry, economic, political, or legal changes;
• exceptions to those exclusions where the target is disproportionately affected;
• language addressing pandemics, war, terrorism, natural disasters, or changes in law;
• a separate closing condition requiring the seller’s representations to remain accurate;
• a bring-down standard that distinguishes fundamental representations from ordinary representations.

The exclusions are not decorative. They decide who bears the risk of an event that affects the target. If a market downturn applies to every participant in the industry, the buyer may have difficulty relying on an MAE. If the target suffers a materially worse impact than its peers, a disproportionate-effects exception may become important.

The buyer should also resist the temptation to use the MAE clause as a substitute for more specific protection. If the concern is the loss of a named customer, the covenant and representations should address that customer directly. If the concern is a pending regulatory approval, it should appear as an express closing condition. If the concern is a minimum level of earnings, the parties may need a financial covenant, a purchase-price adjustment, or a termination right with defined thresholds.

A court is more likely to take a carefully negotiated allocation of risk seriously than a vague argument that the target no longer resembles the business the buyer hoped to acquire.

Working capital adjustment disputes are rarely about one number

The working capital adjustment is another place where the SPA turns business assumptions into litigation risk.

In a closing-accounts structure, the buyer typically pays an estimated purchase price at closing and later calculates the final amount using actual cash, debt, and working capital. The working capital target is intended to leave the buyer with enough operating assets and liabilities to run the business in the ordinary course. That sounds simple until the parties disagree about what “ordinary course” means in the accounting schedule.

A working capital adjustment dispute may involve:

• whether certain receivables are collectible;
• how obsolete or slow-moving inventory is reserved;
• whether customer deposits are treated as working capital or debt-like items;
• how deferred revenue is classified;
• whether transaction-related bonuses belong in working capital;
• whether the seller changed collection or payment practices before closing;
• whether the target was delivered with a normal seasonal balance;
• whether the accounting principles in the SPA override historical practice.

The key drafting issue is hierarchy. The agreement should identify which accounting principles control if they conflict: the specific illustrative calculation, the agreed accounting methods, consistently applied historical practices, or generally accepted accounting principles. “GAAP” alone may not resolve the dispute because several treatments can be technically defensible while producing different results.

The parties should also agree on the process. Who prepares the closing statement? How long does the other party have to object? What level of detail must an objection contain? Can the buyer introduce a new accounting theory after the objection period? Is the independent accountant deciding the amount of the adjustment or interpreting the contract? These questions determine whether the post-closing process is a calculation or a second negotiation.

Indemnification limits in an SPA do not necessarily solve a working capital dispute. A purchase-price adjustment is often treated as a direct calculation of what the buyer paid, while indemnification addresses losses arising from breaches or specified liabilities. The agreement should say how the mechanisms interact and prevent double recovery.

Strategic Enforcement: When Specific Performance Saves the Deal

Most buyers want the right to recover money if a seller breaches. In some transactions, that remedy is not enough. The target may be unique, the market may have moved, or the buyer may have spent years building a strategy around the acquisition. A damages claim after the seller has sold to someone else does not deliver the business.

That is where specific performance becomes strategically important. The remedy can require a party to perform its contractual obligations, including taking steps necessary to close, subject to the terms of the agreement and the court’s equitable powers. It is not automatic. The SPA must authorize it clearly, and the party seeking the remedy must still satisfy the applicable legal standard.

A buyer seeking specific performance generally needs to show more than disappointment. The claim may depend on proving that:

• the agreement is valid and enforceable;
• the relevant closing conditions have been satisfied or would have been satisfied;
• the buyer is ready, willing, and able to close;
• the seller’s breach is material;
• monetary damages are inadequate;
• the contract’s remedy language supports equitable relief;
• the requested order is sufficiently specific for a court to enforce.

This is why closing mechanics matter before a dispute begins. A buyer that cannot demonstrate committed financing, delivered notices, compliance with its own covenants, and readiness to fund may find its enforcement position weakened. The seller, meanwhile, may argue that the buyer’s conditions were not met or that the buyer is attempting to force a closing without satisfying negotiated protections.

Remedy provisions should therefore be negotiated alongside the conditions to closing, not added at the end as boilerplate. The parties should consider whether specific performance is available to either side, whether there are caps or carve-outs, whether the buyer must pursue damages first, and whether an express election of remedies applies. A reverse termination fee may be the seller’s preferred protection, but it should not be drafted in a way that unintentionally gives the buyer only a damages remedy when the commercial objective is to acquire the company.

Enforcement is also a credibility exercise

A demand for specific performance is expensive, urgent, and fact-intensive. The buyer will be asking a court to preserve a transaction while the business, employees, customers, financing, and competing opportunities continue to move. The strongest position is built before the complaint is filed.

That means preserving a clear record of:

• the seller’s obligations under the SPA;
• the buyer’s timely requests for information and consent;
• the satisfaction of closing conditions;
• financing availability and funds flow;
• communications showing the seller’s refusal or attempted breach;
• any third-party sale process that threatens the agreed transaction;
• the steps the buyer took to remain ready to close.

The seller faces a parallel discipline. If it believes the buyer has breached a covenant or failed a condition, it should identify the contractual basis precisely. General claims that the buyer “changed its mind” or “was not cooperative” are weaker than evidence tied to a defined closing condition, notice requirement, or representation.

A broken deal often produces a strange reversal of perspective. The buyer may have entered the process focused on acquiring the business and left focused on enforcing a document. The seller may have treated the SPA as the final administrative step and discovered that its interim conduct created the central legal issue. In both cases, the remedy is shaped by the drafting long before either party reaches court.

What a Broken Deal Teaches About the SPA

The most valuable lesson is not that every transaction needs a longer agreement. It is that the agreement needs to reflect the disputes the parties are most likely to have.

If the price depends on adjusted EBITDA, define the adjustments and the accounting treatment with enough precision to survive a disagreement. If the buyer is concerned about interim operations, draft the ordinary-course covenant around the target’s actual business rather than copying a generic list. If a customer, license, permit, or financing commitment is essential, address it through a specific representation or closing condition. If the buyer may need the company itself rather than a damages award, negotiate specific-performance rights as part of the commercial bargain.

The SPA is not a substitute for trust. Nor is it evidence that the parties distrust each other. It is the place where trust is converted into observable obligations and where uncertainty is assigned before it becomes expensive.

The one-third LOI failure figure is therefore best understood as a warning about the distance between commercial enthusiasm and a completed transaction. Axial’s failed-deal data adds detail to that warning, particularly around diligence and QoE discrepancies, but it should not be misread as a closing-rate calculation. A failed deal is not merely a lost opportunity. It is evidence of which assumptions were never truly agreed.

The parties that learn from that evidence do not necessarily produce the longest SPA. They produce the clearest one: an agreement that says what the business must look like, how it may be run before closing, how the price will be calculated, what happens when the numbers move, and what the buyer can do when the seller no longer wants to perform.

That is the real work of an SPA in mergers and acquisitions. The document does not make a deal certain. It makes the remaining uncertainty legible—and gives the parties a better chance of surviving it.