An indemnity is a contractual promise by one party to cover another party’s financial losses from a specified event, claim, or transaction. The party making the promise is the indemnitor; the party receiving the protection is the indemnitee. If the defined risk materializes, the indemnitor pays. That is the whole mechanism, and it shows up in insurance policies, software licenses, construction contracts, commercial leases, and business acquisitions.
What trips people up is assuming the protection is broader than it is. An indemnity only reaches what the contract defines. A clause that covers third-party injury claims does nothing about breach-of-contract losses unless the language says so. Everything else about indemnity, including whether it is enforceable and how much it will actually pay, is a question about the words on the page.
Indemnify, Defend, and Hold Harmless
Contracts rarely use “indemnify” by itself. The usual phrasing is “indemnify, defend, and hold harmless,” and each term does something distinct.
Indemnify is the reimbursement promise. The indemnitor pays the indemnitee’s established losses after those losses are known. Defend is an earlier obligation: pay for the indemnitee’s legal defense when a covered claim is filed or threatened, before anyone knows whether the claim has merit. Hold harmless aims to prevent the indemnitee from being held legally responsible in the first place.
The practical gap between “defend” and “indemnify” is the one that catches people. The duty to defend kicks in sooner and reaches further. An indemnitor can owe a legal defense for a claim that ultimately turns out to be meritless, meaning no indemnity payment is ever owed but the defense bill still had to be paid. The duty to indemnify only ripens after liability and damages are actually determined.
“Hold harmless” is treated by many courts as a synonym for “indemnify” when the two appear together. On its own, it may be read more narrowly. Some jurisdictions interpret a standalone hold-harmless clause as an agreement not to sue rather than a promise to reimburse. The safer drafting uses all three terms when full protection is the goal.
One-Way vs. Mutual Indemnity
Indemnity can run in one direction or both. A one-way clause means only one party agrees to cover the other’s losses, which is common when the parties have unequal bargaining power or when one side carries far more risk. A manufacturer selling through a distributor might require the distributor to indemnify the manufacturer for losses tied to how the distributor markets or handles the product.
A mutual clause means each party indemnifies the other for losses caused by its own actions. Construction contracts often use mutual indemnification because both the property owner and the contractor face exposure to injury claims, property damage, and regulatory violations during the project. Which structure fits depends on negotiating leverage and how risk is actually distributed.
The Three Forms of Indemnity
Not every indemnity clause transfers the same amount of risk. The form controls how much.
- Broad form. The indemnitor covers all losses regardless of fault. Even if the indemnitee is entirely at fault, the indemnitor pays. This is the most aggressive form and the one most frequently restricted by state statute.
- Intermediate form. The indemnitor covers losses as long as the indemnitee isn’t solely at fault. If the indemnitee is 99% responsible and the indemnitor is 1% responsible, the indemnitor still pays for everything. Only pure fault on the indemnitee’s side releases the obligation.
- Limited form. Each party pays for its own share of fault. If the indemnitor is 60% at fault and the indemnitee is 40%, the indemnitor covers 60%. This is the most balanced form, and arguably not true indemnity at all, since neither party is absorbing the other’s liability.
The language signals which form applies. “In whole or in part” points to broad form. “Caused in part” typically signals intermediate form. “To the extent of” usually means limited form. These phrases are easy to skim past and dramatically change who pays when something goes wrong.
Where Indemnity Shows Up
Insurance Policies
Insurance is indemnity in its most familiar form. The insurer promises to compensate the policyholder for covered losses in exchange for premiums. Auto insurance covers accident-related damages and legal fees. Professional liability policies, like medical malpractice or errors-and-omissions coverage, protect professionals from claims arising out of their work. The insurer is the indemnitor; the policyholder is the indemnitee. Once an insurer pays a claim, subrogation lets it step into the policyholder’s shoes and pursue the person who actually caused the loss.
Business and Software Contracts
In service agreements and software licenses, indemnity clauses allocate the risk of third-party claims. A software vendor typically indemnifies the customer against claims that the software infringes someone else’s patent or copyright. The customer, in turn, often indemnifies the vendor for losses caused by misuse of the software or breach of the agreement.
Intellectual property indemnity clauses usually come with specific remedies if an infringement claim succeeds: the vendor may be required to obtain a license so the customer can keep using the software, replace it with a non-infringing version, or issue a refund if neither option is workable. These clauses also carve out situations where the vendor isn’t responsible, such as infringement caused by the customer modifying the software or combining it with other products in unauthorized ways.
Real Estate and Construction
In real estate transactions, a buyer often seeks indemnity from the seller for problems that existed before the sale but weren’t disclosed, such as environmental contamination, structural defects, or title disputes. Construction contracts contain some of the most heavily negotiated indemnity provisions in any industry, with a contractor typically indemnifying the property owner against injury claims and property damage during the project. The scope of that promise swings widely depending on which of the three forms the contract uses.
Business Acquisitions
Mergers and acquisitions almost always include detailed indemnity provisions. The seller indemnifies the buyer against losses from problems that predate the sale, including undisclosed debts, pending lawsuits, tax liabilities, and environmental issues. The point is to keep the buyer from paying full price for a business whose hidden liabilities eat into its real value.
Clause Mechanics That Change Your Exposure
Notice Requirements
Almost every indemnity clause requires the indemnitee to notify the indemnitor promptly after learning of a covered claim. This isn’t a formality. Late notice can reduce or eliminate the indemnitor’s obligation, particularly when the delay deprived the indemnitor of its right to control the defense. Some jurisdictions require the indemnitor to show actual prejudice from the delay before the obligation is excused. Others treat late notice alone as enough to forfeit the claim.
Caps, Baskets, and Deductibles
Indemnity obligations are almost never unlimited. Contracts constrain them in several ways. A cap sets the maximum total the indemnitor will ever pay; in acquisitions, a common cap for general claims is around 10% of the purchase price, though claims involving fraud are often uncapped. A deductible means the indemnitor pays nothing until aggregate claims exceed a stated threshold, with the indemnitee absorbing everything below it. A tipping basket works like a deductible with one critical difference: once claims cross the threshold, the indemnitee recovers everything from dollar one, not just the excess.
Limitation-of-liability clauses interact directly with indemnity. A contract might grant broad indemnity rights while capping total liability at the fees paid under the agreement. The indemnity technically covers a wide range of losses, but the payout can never exceed the cap. These provisions are usually negotiated together rather than in isolation.
Consequential Damages
Many commercial contracts exclude consequential damages from indemnity obligations. Direct damages are the immediate, natural result of a breach. Consequential damages are the downstream losses that flow indirectly, like lost profits, lost business opportunities, or reputational harm. A mutual waiver of consequential damages can significantly reduce indemnity exposure, but sophisticated parties often carve out an exception for third-party claims, so the indemnity still covers outside lawsuits even when those lawsuits seek consequential damages.
Survival Periods
Indemnity rights don’t last forever. A survival clause sets the window during which the indemnitee can bring a claim after the contract ends or a deal closes. For general representations and warranties in acquisitions, 12 to 18 months is the market standard. Fundamental representations, such as ownership of the company’s assets or proper organization, often survive for five to six years or the applicable statute of limitations, whichever is shorter. Fraud claims usually survive indefinitely. Once the window closes, the right to claim indemnity is gone, regardless of what the governing statute of limitations would otherwise allow.
When an Indemnity Clause May Not Hold Up
Anti-Indemnity Statutes
Roughly 45 states have enacted anti-indemnity statutes, and nearly all of them target the construction industry. These laws exist to prevent parties with superior bargaining power from forcing subcontractors to absorb risks they cannot control. The most common prohibition is against broad-form indemnity, where the subcontractor would be on the hook even for injuries caused entirely by the general contractor’s or owner’s own negligence.
Approaches vary. Some states only bar indemnity for the indemnitee’s sole negligence. Others prohibit indemnity for any degree of the indemnitee’s own fault. A handful have no anti-indemnity statute at all, and even in those states, courts tend to read broad-form clauses narrowly and will not enforce them unless the intent to indemnify for the other party’s negligence is spelled out in unmistakable language.
Unconscionability
Outside construction, courts can strike down indemnity clauses as unconscionable. This requires more than showing the clause is one-sided. The challenger has to show that enforcing it would shock the conscience, typically because a large company used boilerplate to force a consumer or small business to absorb risks they had no realistic ability to negotiate around. Courts look at both the process (was there meaningful bargaining?) and the substance (is the result fundamentally unjust?).
What to Check Before Signing
A few things in an indemnity clause deserve close attention before you agree to it.
Direction. Is the clause mutual or one-way? A clause that only protects the other party is a warning sign unless there’s a clear reason you should carry all the risk.
Cap. Is there one? An uncapped indemnity means theoretically unlimited exposure, which is a serious commitment in any commercial relationship.
Trigger language. An indemnity that covers losses “arising out of or related to” the agreement is far broader than one limited to losses “caused by” a specific party’s breach or negligence. The preposition does real work.
Defense obligation. If the clause includes a duty to defend on top of a duty to indemnify, you could end up funding someone else’s legal defense for a claim that ultimately goes nowhere.
Survival. An indemnity that survives indefinitely is a long tail of risk that follows you well after the business relationship ends. The window should match the kind of exposure being allocated.