Term Sheet vs Letter of Intent: What Is Binding and What Is Not

A term sheet and a letter of intent (LOI) are preliminary deal documents that summarize the economic and structural terms of a proposed transaction before the definitive agreement is signed. Both are usually labeled "non-binding," but that label is misleading: specific provisions inside them — exclusivity, confidentiality, expense allocation, governing law, and any express commitment to negotiate in good faith — are almost always enforceable. Getting that carve-out right is the difference between a walk-away option and a nine-figure judgment.

Founders and operators sign term sheets and LOIs all the time — for M&A, venture financings, real estate deals, and commercial partnerships — and most sign them believing "non-binding" means "no exposure." It does not. Courts in Delaware, New York, and Texas have repeatedly enforced pieces of documents the parties assumed were preliminary. This guide walks through what each document is, which clauses actually bind you, the case law every operator should know, and a step-by-step drafting checklist you can lift into your next deal.

Term Sheet vs Letter of Intent: The Real Difference

In practice, the two documents overlap heavily. Both capture headline economics and structural terms before lawyers spend real money drafting the purchase agreement. The differences are mostly format and posture.

Term sheet. A bullet-style document listing deal terms (price, structure, closing conditions, governance, protective provisions). Common in venture financings — the NVCA Model Term Sheet is the market standard and was refreshed on October 2, 2025 to address tranched financings, national-security compliance, and updated corporate-governance provisions.
Letter of intent. A narrative letter, buyer-to-seller, expressing the buyer's intent to acquire and setting out the same headline terms. Standard in M&A after an initial term sheet has aligned the parties, and usually the trigger for full-scope diligence and drafting the definitive agreement.

Sequence matters. In most middle-market M&A deals, a short term sheet lands first to agree on price and structure. Once the parties are aligned, the buyer issues a more formal LOI that moves the deal into 45–120 days of exclusive diligence. Both documents typically declare themselves non-binding — and both typically carve out several provisions that bind anyway.

Takeaway: Do not choose between "term sheet" and "LOI" based on the label. Choose based on which binding provisions you need enforced today, because those clauses will survive whichever wrapper you use.

What Is Actually Binding in a Term Sheet or LOI

The default drafting convention is: the document as a whole is non-binding, except for a specific list of clauses that are expressly stated to be binding. Those carve-outs are the leverage points. Five provisions almost always bind:

Exclusivity / no-shop. The seller cannot solicit, entertain, or continue discussions with other buyers for a defined window — typically 30 to 120 days. Under New York and Delaware law, a "non-binding" cover paragraph does not defeat this clause; if it is supported by consideration (usually the buyer's commitment to spend on diligence), it is enforceable and breach exposes the seller to damages.
Confidentiality. Financials, customer lists, employee data, and IP disclosed during diligence stay confidential. Often layered on top of a separate NDA. Breach can support both damages and injunctive relief.
Expense allocation. Who pays legal, accounting, and diligence fees if the deal collapses. In venture deals a "broken deal fee" is often folded into the closing cost stack; in M&A the parties usually bear their own.
Governing law and forum. Delaware is the default for VC financings and mid-to-large M&A. New York is common for cross-border. The choice is binding and controls how every other carve-out is interpreted.
Express commitment to negotiate in good faith. This is the trap most operators miss. If the document says the parties will negotiate "in good faith" toward a definitive agreement consistent with the term sheet, Delaware courts have held that obligation is independently enforceable — and the remedy can be expectation damages, not just reliance costs.

Everything else — purchase price, closing conditions, reps and warranties, indemnity caps — is usually non-binding until it lands in the signed definitive agreement.

Takeaway: Read the "Binding Effect" section first, not last. If exclusivity, expense allocation, or good-faith negotiation is buried in the binding block and you did not intend to lock in, negotiate them out or add an outside date and specific carve-backs before you sign.

Three Cases Every Operator Should Know

Pennzoil v. Texaco (1985): $10.53 billion for interfering with an "agreement in principle"

In January 1984, Pennzoil and Getty Oil signed a Memorandum of Understanding at $110 a share and issued joint press releases announcing an "agreement in principle." The next day, Getty's board accepted a higher offer from Texaco at $128 a share. Pennzoil sued in Texas for tortious interference. A jury awarded $10.53 billion — at the time the largest civil verdict in U.S. history — and Texaco eventually settled for $3 billion in 1987, on the way to Chapter 11. The MOU was called preliminary. It was treated as a contract.

SIGA Technologies v. PharmAthene (Delaware Supreme Court, 2013): non-binding term sheet, binding good-faith obligation

SIGA and PharmAthene signed a license term sheet marked "non-binding" that set profit splits, upfront payments of $6 million, and milestone payments of $10 million. When the underlying vaccine looked more valuable than expected, SIGA came back with a materially worse deal — upfront payments of $100 million, milestones of $235 million, and different economics on the profit split. The Delaware Supreme Court held that an express contractual obligation to negotiate in good faith consistent with a term sheet is enforceable even when the term sheet says it is non-binding, and awarded expectation damages. The "non-binding" label did not save SIGA.

Twitter v. Musk (Delaware Chancery, 2022): specific performance turned an option into an obligation

In April 2022, Elon Musk agreed to acquire Twitter for $54.20 a share, roughly $44 billion. The merger agreement included a reciprocal $1 billion termination fee and — critically — a specific performance clause. When Musk tried to walk in July 2022, Twitter sued in Delaware Chancery. The specific performance clause meant Twitter could ask the court to force Musk to close rather than accept the $1 billion. Days before trial in October 2022, Musk closed at the original $54.20. The $1 billion break fee was structurally irrelevant because a court could compel closing.

Takeaway: Assume the counterparty's lawyers will read your preliminary document the way SIGA's did — for enforceable hooks. Any language committing you to negotiate in good faith, or granting the other side specific performance, converts a "preliminary" document into a real obligation. Word it deliberately or delete it.

Binding vs Non-Binding: A Clause-by-Clause Cheat Sheet

Use this as a first read on any term sheet or LOI that crosses your desk. Default classifications; always confirm against the document's own "Binding Effect" section.

Purchase price / valuation: non-binding. Subject to diligence and definitive agreement.
Deal structure (stock vs asset, cash vs stock): non-binding, but usually anchoring.
Closing conditions / MAC: non-binding as drafted in the LOI; binding once in the definitive agreement.
Exclusivity / no-shop: binding. Typical windows: 30–60 days for smaller deals, 60–120 days for mid-market M&A, 45–90 days for VC financings.
Confidentiality: binding. Often survives termination of the LOI by 1–2 years.
Expense allocation / break fee: binding. In signed M&A definitive agreements, break fees typically run 2–4% of deal value.
Reverse termination fee: binding once in the definitive agreement — but check for a specific-performance carve-out that can override it (see Musk-Twitter).
Governing law / jurisdiction: binding.
Good faith negotiation: binding if expressly stated (see SIGA). Delete or scope it if you want an actual walk-away.
Publicity / press release: often binding, and worth reading — see Pennzoil.
Reps and warranties, indemnities, escrow: non-binding. Deferred to definitive.

Takeaway: Before signing, mark each clause "B" or "NB" in the margin and reconcile with the Binding Effect section. Any mismatch is a redline.

A Step-by-Step Drafting and Negotiation Checklist

Decide which carve-outs you need. Buyer usually wants exclusivity and confidentiality binding. Seller wants confidentiality and expense allocation binding, and often pushes back on good-faith language.
Put every binding clause in a single "Binding Provisions" block. Enumerate them explicitly. Do not sprinkle "the parties shall…" across the document. Silent binding language is what created SIGA v. PharmAthene.
Give exclusivity an outside date and a fiduciary out. Public-company sellers typically negotiate a "superior proposal" carve-out; private sellers should at minimum insist on a hard end date after which the no-shop lapses automatically.
Cap expense allocation. If you agree to reimburse the buyer's diligence costs on breakup, cap it (e.g., $150K for a middle-market deal) and require documented invoices.
Handle "good faith" deliberately. If you want to walk if diligence turns up problems, do not commit to negotiate a definitive agreement "consistent with the terms of this LOI in good faith." Use "the parties intend to proceed to negotiate" without the express duty.
Pick the governing law before you argue price. Delaware for U.S. venture and M&A default; New York for cross-border. The choice controls how ambiguous carve-outs are read.
Coordinate press releases. Any public "agreement in principle" statement can be dangerous. Silence, or a joint statement drafted by counsel, is safer.
Sunset the whole document. Add a termination clause: if a definitive agreement is not signed by day X, the LOI terminates in full and only the confidentiality and expense provisions survive.

Takeaway: Treat the term sheet or LOI as the first negotiation, not a formality. The clauses you sign here determine what leverage you have in the last week before closing.

Conclusion: Preliminary Is Not the Same as Optional

The phrase "non-binding" on the front of a term sheet or LOI is a starting position, not a legal conclusion. Pennzoil turned a memorandum of understanding into a $10.53 billion jury verdict. SIGA turned an express "non-binding" term sheet into expectation damages under Delaware law. Twitter turned a signed merger agreement's specific-performance clause into a court order forcing a $44 billion close. In each case, the parties who lost had signed something they thought was preliminary — and every time, the enforceable pieces were exactly the ones a careful reader would have flagged before signing.

The right defense is boring: a clean, versioned template with the Binding Effect section drafted first, a checklist of enforceable carve-outs, and a redline protocol so nothing sneaks past. That is exactly what ModelStack's IB and M&A template kit ships: a market-standard LOI, a term sheet for venture financings aligned to the 2025 NVCA update, an exclusivity clause library with negotiated fallbacks, and a definitive-agreement mapping so the same defined terms carry through to signing. Buy it once, use it on every deal, and stop paying to redraft the same six clauses at $700 an hour.