The practical answer to the question of what is a non-compete agreement is simple: it is a contract clause that limits where a worker can go after leaving a job, usually by blocking work for a competitor, starting a competing business, or both. In the U.S., that clause only matters if the drafting is narrow enough and the applicable state law allows it to be enforced. This article breaks down how these agreements work, what they usually cover, what makes them risky, and what both sides should check before they sign.
The short version is that enforceability depends on state law and the clause itself
- Non-competes restrict post-employment work, but they are not the same as NDAs or non-solicitation clauses.
- As of 2026, the FTC’s nationwide non-compete rule is not in effect and not enforceable, so state law still drives most outcomes.
- California generally voids most employment non-competes, and Minnesota voids most post-termination non-competes with narrow exceptions.
- The three biggest drafting issues are duration, geography, and the exact kind of work the clause tries to block.
- In practice, the disputes that matter most are about trade secrets, client relationships, and whether the restriction is broader than necessary.
What a non-compete agreement actually restricts
A non-compete is a restrictive covenant. In plain English, it says a person cannot do certain competitive work for a set time after employment ends, and sometimes cannot start a competing business at all. Employers use them to try to protect customer relationships, confidential information, and specialized know-how that would be hard to replace quickly.
I usually separate a non-compete from the contracts that sit beside it. A non-disclosure agreement limits the use or sharing of confidential information. A non-solicitation clause usually bars a former worker from poaching clients or coworkers. A non-compete goes further because it tries to stop the person from competing in the first place, which is why it draws more legal scrutiny.
| Restriction | What it limits | Why employers use it |
|---|---|---|
| Non-compete | Working for a competitor or starting a competing business after leaving | Protects market position, but often faces the toughest legal challenge |
| Non-solicitation | Contacting customers, vendors, or coworkers for business purposes | Targets relationships instead of banning all competition |
| NDA | Using or disclosing confidential information | Protects trade secrets and internal data |
| Garden leave | Immediate job switching during a paid notice period | Buys transition time without a hard post-employment ban |
That distinction matters because courts often treat a narrow NDA differently from a broad restraint on future work, and the next question is whether the non-compete survives the legal test where the worker lives and works.
How enforceability works in the U.S. in 2026
As of 2026, the federal picture is less sweeping than many people assume. The FTC adopted a nationwide non-compete rule in 2024, but the agency’s own rule page says it is not in effect and not enforceable after a court order and the FTC’s later decision to dismiss its appeal. In practice, that leaves state law, state courts, and contract drafting as the real decision-makers.
That is why outcomes vary so much. California Business and Professions Code section 16600 broadly voids contracts that restrain someone from engaging in a lawful profession, trade, or business, and it explicitly reads broadly against employment non-competes. Minnesota’s statute is also blunt: most covenants not to compete are void and unenforceable, with narrow exceptions tied to the sale or dissolution of a business.
In states that still allow employee non-competes, courts usually ask whether the clause is reasonable in scope, time, and geography, and whether it protects a legitimate business interest rather than just suppressing movement. A one-year restriction on a very specific sales role may survive where a three-year, nationwide ban on any role in a large industry would not.
I also watch for states that let judges trim overbroad language and states that refuse to rescue bad drafting. That difference can decide whether a clause is narrowed or thrown out completely, which is why the next layer of analysis is always the wording itself.
The fine print that changes the risk
Two non-competes can look similar and still have very different outcomes. The details below usually tell me more than the label on the first page.
| Clause element | What to check | Why it matters |
|---|---|---|
| Time limit | How long the restriction lasts after departure | Shorter periods are easier to defend; longer periods are harder to justify |
| Geographic scope | Whether it is local, regional, national, or tied to a customer footprint | A radius that matches real business territory is more defensible than a blanket ban |
| Activity scope | Whether it blocks only direct competitors or any similar work | Overbroad job definitions can turn a narrow protection into a career blocker |
| Trigger event | Whether the restriction starts after resignation, termination, layoff, or severance | The business impact is different if the clause applies after a firing versus a voluntary exit |
| Consideration | What the employee got in exchange for signing | Some states care whether the promise was supported by real value, especially when the clause is added later |
| Choice of law and venue | Which state law governs and where disputes must be heard | These clauses can shift leverage, but they do not always override the worker’s home-state protections |
The red flags are usually easy to spot: broad industry bans, long durations, vague “similar business” language, or restrictions that follow the employee far beyond the market they actually served. When I see those, I assume the clause will be disputed unless the employer has a very specific justification.
What happens when someone leaves a job
Departure is where the clause stops being theoretical. An employer may send a cease-and-desist letter, ask for a written acknowledgment, or go straight to court and seek an injunction, which is a fast order meant to stop the new job from moving forward. Some cases settle quickly because neither side wants to spend money litigating a clause that may be narrower on paper than it is in practice.
For employees, the practical risk is not only a lawsuit. A new employer may back away if the clause looks messy, investors may want a legal opinion, and a poorly handled transition can create a trade-secret problem even if the non-compete itself is weak. That is why I tell people to separate the contract fight from the conduct fight: you can sometimes beat a non-compete and still lose if you walked out with confidential files or client lists.
- Read the clause before you resign, not after.
- Compare the new role to the exact restricted activities, not the job title.
- Keep evidence that you returned devices, files, and access credentials.
- Ask for a release or written carve-out if the new role is close but not identical.
- Do not assume “unenforceable” means “safe to ignore.”
The next question is obvious: if a broad non-compete is hard to defend, what should a business use instead?
Better ways to protect a business without overreaching
From a strategy standpoint, I rarely think a company should lead with a sweeping non-compete unless the risk profile is truly unusual. In many businesses, narrower tools do the job better and with less litigation risk.
- Confidentiality agreements protect trade secrets, pricing, source material, and internal plans.
- Non-solicitation clauses stop former workers from raiding clients or employees, which is often the real business loss.
- Garden leave keeps the employee on payroll for a notice period and buys time without a hard post-employment ban.
- Access controls and data segmentation reduce exposure before someone leaves, which matters more than many legal documents do.
- Targeted retention or bonus plans can protect key relationships without pretending the worker has no right to change jobs.
These tools are not magic. A weak NDA will not stop misuse of information, and a sloppy non-solicit can be attacked for the same reasons a non-compete can. But they are often easier to defend because they target actual harm rather than trying to block competition wholesale.
The practical checklist I use before anyone signs
If I am reviewing a non-compete today, I start with five questions: Which state law applies? What exact behavior is prohibited? How long does the clause last? What business interest is it really protecting? And is there a narrower tool that would cover the same risk?
That sequence usually gets to the truth faster than reading the contract line by line in isolation. A well-drafted restriction is focused, tied to a real business need, and consistent with local law; a weak one tries to do too much and often creates more leverage for negotiation than protection for the employer. If the clause sits in that gray area, the safest move is to slow down, get the governing law right, and treat the agreement as a business decision as much as a legal one.
