Non-compete clauses are one of the most overused and least understood tools in Australian employment law. Startups routinely include them in employment contracts — often copied from a template or borrowed from a larger company’s playbook — without understanding what they actually protect, whether they’re enforceable, or how the law is about to change.
The result is a false sense of security. A founder assumes their employment agreement prevents a departing engineer from joining a competitor, only to discover — when it actually matters — that the clause is too broad, too long, or too vaguely drafted to hold up.
Here’s what Australian startups actually need to know about non-competes after employment.
The Starting Position: Restraints Are Presumed Void
Under Australian common law, a post-employment non-compete clause is a restraint of trade. The default position is that restraints of trade are void and unenforceable as contrary to public policy. The rationale is straightforward: people have a right to earn a living, and the law is reluctant to restrict that right.
The burden falls on the employer to prove that the restraint is reasonably necessary to protect a legitimate business interest. If the employer can’t clear that bar, the clause is unenforceable — regardless of what the employee agreed to when they signed their contract.
This is worth emphasising because many founders assume that because an employee signed an agreement containing a non-compete, the clause is binding. It isn’t — not automatically. Signature doesn’t equal enforceability.
What Courts Actually Look At
When a court assesses the reasonableness of a non-compete, it considers several factors:
The interest being protected. A non-compete must protect a legitimate business interest. The three categories Australian courts consistently recognise are: trade secrets and confidential information, client relationships and connections, and the stability of the workforce. A clause that doesn’t connect to one of these interests won’t be enforced. “We don’t want them working for a competitor” is not, by itself, a legitimate interest.
Duration. How long does the restraint last? A 3-month restraint for a junior marketing hire is more likely to survive than a 24-month restraint for the same role. Courts assess duration relative to the sensitivity of the information and relationships involved.
Geographic scope. Does the restraint cover a suburb, a city, the state, or the whole country? The scope needs to match the actual reach of the business interest. A startup that operates exclusively in Sydney will struggle to enforce a restraint that covers all of Australia.
Scope of restricted activity. What is the employee actually prevented from doing? A clause that prevents someone from working “in any capacity” for any business that could be described as a competitor is far more likely to fail than one that prevents them from performing a specific role in a directly competing business.
The employee’s seniority and access. Courts draw a clear distinction between senior employees with access to confidential information, client relationships, and strategic knowledge, and junior employees who don’t have that level of exposure. A non-compete is far more likely to be enforced against a CTO who built the core product than against a customer support representative.
The NSW Difference
If your startup is based in New South Wales — and most Australian startups are — there’s an important statutory overlay. The Restraints of Trade Act 1976 (NSW) gives courts the power to read down an unreasonable restraint to make it reasonable, rather than striking it out entirely.
In the rest of Australia, courts apply the “blue pencil” test: they can sever unreasonable parts of a restraint clause, but they can’t rewrite it. If the clause is fundamentally unreasonable, it fails entirely.
In NSW, the court has broader discretion. It can modify the restraint — reducing its duration or narrowing its geographic scope — to arrive at something reasonable. This makes restraints in NSW somewhat more employer-friendly than in other states. Research suggests that around 56% of post-employment restraints in NSW are enforced, compared to roughly 33% in the rest of Australia.
This is one reason why well-advised startups use cascading restraint clauses — clauses that specify a series of progressively narrowing alternatives (for example, 24 months or 12 months or 6 months; within Australia or within NSW or within Sydney). The cascading structure gives the court a range of options and increases the likelihood that at least one combination will be upheld as reasonable.
The 2027 Reforms: A Significant Shift
The landscape is changing. In the 2025–26 Federal Budget, the Albanese Government announced that non-compete clauses will be banned for workers earning below the high-income threshold under the Fair Work Act 2009 (Cth) — currently $175,000 per year (excluding superannuation). The reforms are expected to take effect from 2027 and will apply prospectively to employment contracts made or varied after the start date.
The same reform package will also ban:
- Wage-fixing agreements between businesses that cap worker pay.
- No-poach agreements between businesses that prevent them from hiring each other’s staff.
The ban on non-competes won’t affect other types of post-employment restraints. Confidentiality clauses, non-solicitation clauses (of both clients and employees), and obligations to protect trade secrets will remain enforceable under the existing common law framework.
For startups, this reform has several practical implications:
- Most of your team will likely fall below the threshold. If you’re an early-stage startup, the majority of your employees — engineers, designers, marketers, ops staff — probably earn less than $175,000. Non-competes for these employees will be unenforceable once the reforms take effect.
- Senior hires may still be subject to non-competes. Your CTO, VP of Engineering, or Head of Sales earning above the threshold can still be subject to a post-employment non-compete — but only if it satisfies the existing common law reasonableness test.
- You need to shift your protective strategy. Relying solely on non-competes was always risky. After 2027, it will be impossible for most of your workforce. The good news is that the alternatives are often more effective anyway.
What Actually Works: Practical Alternatives
Even without non-competes, startups have several tools to protect their business when employees leave.
Confidentiality Agreements
A well-drafted confidentiality clause is more targeted and more enforceable than a non-compete. It protects the specific information you care about — product roadmaps, pricing models, customer data, proprietary algorithms — without preventing the employee from working. Confidentiality obligations should survive the employment relationship indefinitely or for a defined period (five years is common for commercial confidential information; trade secrets should be protected for as long as they remain secret).
Non-Solicitation Clauses
A non-solicitation clause prevents a departing employee from approaching your clients or poaching your staff for a defined period after leaving. These are narrower than non-competes and more likely to be enforced because they target a specific harm (loss of client relationships or workforce stability) rather than broadly restricting the employee’s ability to work.
IP Assignment
For startups, the most critical asset is usually intellectual property. An IP assignment clause in the employment agreement — confirming that all IP created by the employee in the course of their employment belongs to the company — is essential. This should be supplemented by a confirmatory IP assignment deed on departure, particularly for technical employees.
Garden Leave
A garden leave clause requires the employee to serve out their notice period at home, on full pay, without performing work or contacting clients. During this period, the employee remains bound by their duties of fidelity and confidentiality. Garden leave gives the company a buffer period to transition client relationships and secure information without needing a post-employment restraint at all. Courts generally view garden leave more favourably than non-competes because the employee continues to be paid.
Drafting Non-Competes That Have a Chance
If you do include non-competes — and there are still situations where they’re appropriate for senior hires above the income threshold — draft them to maximise enforceability:
- Use cascading clauses. Specify alternative durations, geographic areas, and scopes of restricted activity so the court can select the most reasonable combination.
- Tailor to the role. Don’t apply the same restraint to every employee. A blanket non-compete across the company signals that the clause isn’t connected to any specific business interest.
- Define the restricted activity precisely. “Working for a competitor” is vague. “Performing software development services for a business that provides [specific type of service] to [specific market]” is defensible.
- Keep durations short. In the startup context, where information becomes stale quickly, 3 to 6 months is realistic. Twelve months is at the outer edge of what courts typically enforce for all but the most senior roles.
- Consider separate consideration. If you’re adding a restraint mid-employment (rather than at hiring), provide something in return — a bonus, a promotion, additional equity. Restraints imposed without fresh consideration are vulnerable to challenge.
The Bottom Line
Non-competes have always been harder to enforce than most founders think. After 2027, they’ll be outright banned for the majority of the startup workforce. The smart move is to build your protective framework around confidentiality, non-solicitation, IP assignment, and garden leave — tools that are more targeted, more enforceable, and more aligned with what you’re actually trying to protect.
If you’re reviewing your employment agreements or preparing for the 2027 reforms, get in touch. We help startups build employment frameworks that actually hold up when they need to.
