You sign things constantly, a supplier order, an employment offer, a SaaS subscription you clicked through without reading. Most never cause trouble, so it is tempting to treat the contract as paperwork. It isn't. A contract is a private law you and the other side write for yourselves, and the day it matters is the day the relationship has already gone wrong. The point of understanding the fundamentals is to read for that day before you sign, not after.

The quick version

  • A binding contract needs a handful of ingredients: an offer, an acceptance of it, consideration (each side gives something of value), an intention to be legally bound, and parties with the capacity and authority to agree.
  • You do not need a signed document for most contracts. A clear offer accepted in exchange for value can bind you, including by email, conduct, or a click.
  • Every contract is incomplete: no document can foresee everything. The clauses that handle "what if it goes wrong", liability, termination, indemnity, dispute resolution, are the ones that decide your real exposure.
  • The fundamentals are general principles, and the detail differs by country. For anything material, have a qualified lawyer in your jurisdiction check it.

What actually makes a contract binding

Lawyers describe a valid contract as a set of elements that must all be present. The exact list varies by source and jurisdiction, but common-law systems converge on the same core: an offer, acceptance, consideration, an intention to create legal relations, and parties with capacity (these are the elements catalogued in standard practitioner explainers such as DocuSign's guide to the essential elements of a contract). Strip out any one and you may have a conversation, a courtesy, or a moral promise, but not something a court will enforce.

Two of these trip people up most. The first is that a contract usually does not require a signature or even a written document. An offer that is accepted, supported by consideration, with intent to be bound, is generally a contract however it was formed. The famous illustration is Carlill v Carbolic Smoke Ball Co (English Court of Appeal, 1893): a company advertised £100 to anyone who used its flu remedy as directed and still caught flu, and even deposited £1,000 to show it meant it. Mrs Carlill did exactly that, caught flu, and claimed. The court held the advertisement was a genuine offer to the world, accepted by performance, a binding contract, no signature in sight. The practical lesson is uncomfortable: your emails, your "yes, go ahead", and your team's conduct can all form a contract. If you are still negotiating, say so in writing, "subject to contract", so a stray message doesn't bind you early.

The second is consideration: each side must give something of value. A bare promise to do someone a favour, with nothing coming back, is generally not enforceable as a contract. This is why agreements sometimes recite a nominal sum, and why a one-sided variation ("we'll cut your price, you give nothing new") can be shaky. When you change a deal, then, make sure both sides give something, even something small, so the change is itself supported by consideration and sticks.

flowchart TD
  A(["Offer
clear, definite terms"]) --> B{"Accepted on
those terms?"}
  B -->|"No, counter-offer"| A
  B -->|"Yes"| C{"Consideration
both sides give value?"}
  C -->|"No"| X(["Not an enforceable
contract"])
  C -->|"Yes"| D{"Intention to be
legally bound?"}
  D -->|"No, e.g. social"| X
  D -->|"Yes"| E{"Capacity & authority
to agree?"}
  E -->|"No"| X
  E -->|"Yes"| F(["Binding contract"])
The formation gate: a deal has to clear every test, not most of them. Leaders Loop

An honest limitation. This checklist is the common-law frame. Civil-law countries (much of Europe, Latin America, parts of Asia) do not require consideration in the same way, and consumer-protection rules can override what the parties agreed. The elements above are a reliable way to think about whether you are bound, not a substitute for advice on the law that actually governs your contract.

Why every contract is incomplete, and what to do about it

One idea changes how you read a contract for good: no document can anticipate everything. The world is too uncertain, and writing down every contingency would cost more than it is worth. Economists call the result the incomplete contract, and it sits at the centre of the field's most decorated work. Oliver Williamson built transaction-cost economics on the recognition that, because people are boundedly rational and sometimes opportunistic, contracts are inevitably incomplete, so the interesting question is how to govern the gaps. In 2016 the Nobel committee gave the economics prize to Oliver Hart and Bengt Holmström "for their contributions to contract theory", recognising precisely this: contracts and ownership are imperfect responses to the fact that you cannot specify or enforce everything in advance.

A contract is not a prediction of the future. It is a set of rules for what happens when the future surprises you.

That theory has a blunt practical payoff. If the body of a contract describes the deal when everything goes to plan, the value of the document is mostly in the clauses that handle the plan failing. Spend your reading time, then, on the "what if it goes wrong" provisions, that is where incompleteness gets resolved, in your favour or against you. The four that decide most of your real exposure:

  • Limitation of liability, the cap on what each side can be made to pay if things go wrong. An uncapped liability clause is the single most expensive line in many contracts.
  • Indemnity, a promise to cover the other party's losses from specified events (often third-party claims). Indemnities can sit outside the liability cap, so they quietly reopen exposure you thought you'd limited.
  • Termination, how, when, and at what cost either side can exit. A deal you can't leave is a different deal from the one you thought you signed.
  • Governing law and dispute resolution, which country's law applies and where (or how) you fight. This decides how expensive and how foreign any disagreement will be.

An honest limitation. Contract theory explains why these clauses matter; it does not tell you what number to accept. A liability cap at one times annual fees might be generous in one market and reckless in another. The principle is durable, read the failure clauses first, but the negotiation is judgement, and for material deals, professional advice.

A worked example

Take a mid-sized retailer, call it Harbourline, signing a one-year contract with a cloud logistics vendor. (Illustrative figures throughout; this is a teaching example, not a real contract.) The annual fee is, say, £120,000. The operations lead reads the service description carefully, likes the uptime promise, and is ready to sign.

A colleague applies the lens above and reads for incompleteness instead. Three things surface. First, the limitation of liability caps the vendor's total liability at one month's fees, about £10,000, so if an outage halts Harbourline's warehouse for a week, the most they can recover is a rounding error against the real loss. Second, an indemnity requires Harbourline to cover the vendor against any third-party claim arising from Harbourline's data, broad enough to swallow risks Harbourline can't see. Third, termination lets the vendor walk on 30 days' notice, but locks Harbourline in for the full year. None of this was hidden; it was simply in the clauses nobody had read.

flowchart LR
  A(["Read the service
description only"]) --> B(["Looks fine,
ready to sign"])
  C(["Read the failure
clauses too"]) --> D(["Liability capped at
~£10k vs £120k deal"])
  C --> E(["Indemnity reopens
capped exposure"])
  C --> F(["Asymmetric exit:
vendor 30 days, you 1 yr"])
  D --> G(["Negotiate before
signing, not after"])
  E --> G
  F --> G
Same contract, two readings, the value is in the clauses that handle things going wrong. Leaders Loop

The fix is not to refuse the deal; it is to negotiate the gaps before signing, push the cap toward the contract value, carve the indemnity down to risks Harbourline actually controls, and make the exit terms symmetric. The number that matters is the gap between the £10,000 cap and the loss a real outage could cause. Closing that gap on paper costs an email exchange now; discovering it after an outage costs the business.

Frequently asked questions

Does a contract have to be in writing to be valid?

Usually not. Most contracts can be formed orally, by email, or by conduct, and are still binding. There are exceptions where a written, signed document is required by law, common examples include sales of land and certain guarantees, and the categories differ by country. The safer habit is to assume your agreements can bind you even without a formal document, and to put anything important in writing for clarity and proof.

What's the difference between a contract being void and being voidable?

A void contract was never valid, the law treats it as if it never existed (for example, an agreement to do something illegal). A voidable contract is valid until one party chooses to cancel it because of a defect in how it was formed, for instance, a party who lacked capacity, or who was misled or pressured. The practical point: "voidable" gives a party a right to walk away, but the contract stands until they exercise it.

If I click "I agree", am I really bound?

Generally yes. A clickwrap agreement, where you tick a box or click "I agree" after being shown the terms, is widely enforced, because clicking is acceptance of an offer in exchange for the service. Terms hidden behind an obscure link that you were never reasonably shown ("browsewrap") are weaker. Either way, the lesson is that a click can form a contract, so the terms behind it are worth a glance, especially the liability and auto-renewal clauses.

Which clause should I read first?

Read the failure clauses before the description of the deal: limitation of liability, indemnity, termination, and governing law. The body of the contract tells you what you get when everything works; these tell you what happens when it doesn't, and that is where your real exposure lives. If you only have ten minutes, spend them there.

Do I always need a lawyer?

Not for a low-value, standard agreement where the downside is small. But the elements and clauses above are general principles, and the law that governs your specific contract, including consumer rules and local variations, can change the answer. For anything material, high-risk, or unfamiliar, a qualified lawyer in the relevant jurisdiction is cheaper than the dispute you avoid.

Related in the Toolkit

Contract fundamentals connect to nearly everything legal a leader touches: the clauses that allocate who owns what often turn on intellectual property, and the way you cap and shift risk in a contract works hand in glove with insurance and risk transfer.

Where to go next