Every commercial relationship carries the seed of a dispute. A missed delivery. A disputed invoice. A partner who reads the same clause the opposite way. When one surfaces, the instinct is to reach for a lawyer and a writ, but going to court is rarely the smart first move. It is the move you take when the cheaper ones have failed. The skill is knowing the ladder of options below litigation, and climbing it on purpose rather than falling up it in anger.

The quick version

  • Litigation is a public, adversarial process where a judge imposes a decision. It is slow, costly, and you hand control of the outcome to a stranger in a robe.
  • Alternative dispute resolution (ADR) is the family of routes that avoid a courtroom: negotiation, mediation (a neutral helps you settle), and arbitration (a private decision-maker rules).
  • Most disputes never reach trial, the great majority settle. UK commercial mediation settles around 92% of cases brought to it, most on the day itself (CEDR, 2023).
  • This is a general explainer, not legal advice. Procedure, costs rules and ADR obligations differ sharply by country, for any real dispute, take advice in the jurisdiction that governs it.

Litigation is a decision you outsource, at a price

When you sue, you are asking the state to settle your disagreement by force of law. A judge (sometimes a jury) hears evidence, applies the law, and imposes a binding outcome that the courts will enforce. That power is the point, and also the problem. You gain certainty of enforcement but lose control of the result. The decision falls to someone who did not build the relationship, cannot split the difference creatively, and is bound to name a winner and a loser. It is public, slow, and expensive too, the bill climbing at every stage, a letter, then filed proceedings, then disclosure of documents, then a contested trial.

Which is why so few disputes actually get there. In civil justice systems, the overwhelming majority of filed cases settle before a judge ever rules, commonly estimated at around 70% of US civil cases and often higher, as parties confront the cost and uncertainty of fighting on (a pattern documented across practitioner and academic surveys, such as this overview of ADR's rise into the mainstream). The trial is the exception, not the rule. So the move is: treat issuing proceedings as a tool of last resort or deliberate pressure, not as "step one." Before you instruct a lawyer to file, ask what a settlement would realistically look like, because statistically, that is where this ends anyway.

The deeper trap is psychological. Once a business has spent six figures on a case, the urge to keep going to "get something back" grows even as the rational case for settling strengthens, a textbook escalation of commitment to a losing course, first formalised by Barry Staw in 1976. Each stage of litigation raises the sunk cost and the emotional stake. So the move is: decide your walk-away number before the costs mount, write it down, and judge it on what is still recoverable, not on what you have already spent, which is gone either way.

An honest limitation. Litigation is sometimes the right answer. You need it when you require a public precedent, an injunction to stop something now, enforcement against a party acting in bad faith, or a binding ruling on a point of law. The argument here is not "never sue", it is "sue on purpose," knowing you are buying enforcement power at the price of cost, time and control.

flowchart TD
  A(["A dispute
surfaces"]) --> B{"Can we still
talk directly?"}
  B -->|"Yes"| C(["Negotiate
cheapest, you keep control"])
  B -->|"Stuck"| D{"Want help reaching
our own deal?"}
  D -->|"Yes"| E(["Mediation
neutral facilitates, non-binding"])
  D -->|"No, need a ruling"| F{"Keep it private
& final?"}
  F -->|"Yes"| G(["Arbitration
private, binding decision"])
  F -->|"No, need court powers"| H(["Litigation
public, binding, enforceable"])
The escalation ladder: control and cost-efficiency are highest at the top, court powers only at the bottom. Leaders Loop

The ADR ladder: negotiation, mediation, arbitration

"Alternative dispute resolution" is an umbrella for everything between a polite phone call and a courtroom. The three rungs differ on one axis above all: who decides the outcome, you, or someone else.

Negotiation is two parties resolving it themselves. It is the cheapest, fastest, most private option, and you keep complete control. The reason it so often fails is that people argue over positions ("I want £50,000") instead of the interests beneath them (cash flow, a damaged reputation, a relationship worth saving). The foundational text here is Roger Fisher and William Ury's Getting to Yes (1981), from the Harvard Negotiation Project, which reframed bargaining around interests, options for mutual gain, and objective criteria, and gave managers the idea of the BATNA, the Best Alternative To a Negotiated Agreement. Your BATNA is what you will do if the talk fails, and it is the only honest measure of your real power. So the move is: before any settlement conversation, write down both BATNAs, yours and theirs. If your alternative to a deal is a two-year lawsuit you might lose, your appetite for a sensible settlement should be high, and naming that on paper stops bravado driving the decision.

Mediation adds a neutral third party who helps the two sides reach their own agreement. Crucially, the mediator does not decide, they shuttle, probe, reality-test, and create space for a deal. It stays private and non-binding until the parties sign, and it works remarkably well. The Centre for Effective Dispute Resolution's Tenth Mediation Audit (2023) put the aggregate settlement rate of UK commercial mediation at 92%, 73% settling on the day, a further 20% shortly after, across a market it estimated saves businesses around £5.9 billion a year in wasted management time, legal fees and lost productivity. So the move is: propose mediation early, in writing, even if you feel you would "win" in court. It costs a fraction of trial, preserves the commercial relationship, and a refusal to mediate can itself count against a party on costs in some jurisdictions.

In litigation a stranger decides who wins. In mediation, you both stay the authors of the outcome, you just hire a guide.

Arbitration is the private cousin of litigation: the parties appoint an arbitrator (or panel) who hears the case and imposes a binding decision, like a judge but behind closed doors. It is common in cross-border commercial contracts because awards are enforceable internationally under the New York Convention (1958), ratified by over 170 countries. You trade the courtroom for privacy, a chosen expert decision-maker, and (sometimes) speed, but you give up the right to decide the outcome, and appeal rights are narrow. So the move is: sort out your dispute-resolution clause when you sign the contract, not when you fall out. Whether a dispute goes to a London court, a Singapore arbitrator, or mediation first is usually decided years earlier by a clause nobody negotiated, make it one you did.

An honest limitation. ADR is not free of teeth or always cheaper. Arbitration can rival litigation on cost, and its narrow appeal rights cut both ways, a bad award is hard to undo. Mediation only works if both sides will engage in good faith; against a party that is simply playing for time, the court's coercive power is the point. The ladder is a default order of preference, not a rule that binds every case.

A worked example

Take a mid-sized software firm, call it Northwind, in a dispute with a manufacturing client over a delayed implementation. (Illustrative figures throughout; this is a teaching example, not a real case.) The client has withheld a final payment of, say, £180,000 and claims £400,000 in losses from the delay. Northwind's founder is furious, convinced they are in the right, and wants to sue for the unpaid fee on Monday.

The general counsel runs the ladder instead. First, the BATNA check: litigation would cost an estimated £150,000–£250,000 in fees over 18 months, with an uncertain result and the client relationship destroyed regardless of outcome. That is a poor alternative, so the appetite to settle should be real, not a sign of weakness. Second, the interests beneath the positions: the client does not actually want £400,000; they want the system working and a reason to defend the delay to their own board. Northwind does not only want £180,000; it wants a reference customer and no public fight. Those interests overlap in a way the positions hide.

flowchart LR
  A(["Sue for the
£180k fee"]) --> B(["~£200k in fees,
18 months, uncertain"])
  B --> C(["Relationship
destroyed either way"])
  D(["Mediate on
the interests"]) --> E(["System fixed +
partial payment"])
  D --> F(["Face-saving story
for both boards"])
  E --> G(["Deal in weeks,
relationship survives"])
  F --> G
Same dispute, two routes, the costly fight versus the deal hiding inside the parties' real interests. Leaders Loop

Northwind proposes mediation. In a single day, a settlement emerges: Northwind fixes the outstanding issues at its own cost over four weeks; the client pays £120,000 of the withheld fee on completion and waives the loss claim; both sign a short reference statement. Nobody got everything. But Northwind banked £120,000 in weeks rather than gambling on £180,000 in two years, kept a customer, and avoided a quarter-million in fees. The number that mattered was never the £180,000 on the invoice, it was the gap between a mediated deal this month and the real, risk-adjusted value of a lawsuit that might never pay out.

Frequently asked questions

What's the difference between mediation and arbitration?

It comes down to who decides. In mediation, a neutral helps the parties negotiate their own settlement, the mediator cannot impose anything, and nothing binds you until you sign. In arbitration, the arbitrator acts like a private judge and hands down a binding decision you must accept. Mediation keeps you in control of the outcome; arbitration gives you finality but takes the decision out of your hands.

If most disputes settle anyway, why not just sue and settle later?

Because the costs and damage accrue along the way. Each stage of litigation, filing, disclosure, witness statements, hearings, adds fees, time and bad blood, and feeds the sunk-cost trap that makes settling harder, not easier. Settling after two years of fighting usually means a worse deal than the one available before the writ. Suing as a tactic can occasionally force a stalled counterparty to the table, but it is an expensive lever to pull first.

Can a refusal to try ADR hurt me even if I win?

In some jurisdictions, yes. Courts in England and Wales, for example, can penalise a party on costs for unreasonably refusing to mediate, and recent reforms allow courts to order parties into ADR in some cases. The general lesson is that being seen to engage constructively, proposing mediation in writing, in good faith, protects you, while a flat refusal can be held against you regardless of the merits. Check the specific rules where your dispute is governed.

Where do I actually decide which route a dispute takes?

Usually long before the dispute exists, in the contract's dispute-resolution clause. That clause can mandate negotiation first, then mediation, then arbitration or a named court ("escalation" or "tiered" clauses), and can fix the governing law and venue. Most people never negotiate it. Treat it as a real term: it decides how expensive, how private, and how foreign any future fight will be.

Is ADR always cheaper than going to court?

Usually, but not guaranteed. Negotiation and mediation are typically far cheaper and faster than trial. Arbitration, however, can approach litigation on cost, you are paying the arbitrator and the venue privately, and its limited appeal rights mean a poor result is hard to overturn. ADR is the sensible default, but "alternative" does not automatically mean "cheap"; weigh each route on the specific dispute.

Related in the Toolkit

Disputes rarely start in a courtroom, they start in a contract, which is why the clauses you read before signing decide so much of what happens when things go wrong (see contract fundamentals), and why the way you cap and shift risk on paper is really a form of insurance and risk transfer.

Where to go next