That makes "can we legally record this?" a real question, not a technicality. The frustrating part is that the answer changes depending on which side of the Tasman (or the equator) you're on, and the internet is full of confident, contradictory advice.
Here's the honest version, market by market.
The one thing that's true everywhere
Across all three countries, recording a call you are a party to sits on very different legal footing from intercepting a call you are not part of. Secretly tapping someone else's conversation is a serious offence more or less everywhere. Recording your own business calls is generally permitted; but how you do it, what you tell people, and what you do with the recording afterward is where the rules diverge.
So the useful mental model isn't "is recording legal?" It's two separate questions:
- Am I allowed to make the recording?
- Have I handled the resulting data properly?
A firm can be perfectly fine on the first and in breach on the second.
Australia: a two-layer system that causes most of the confusion
Australia has federal law under the Telecommunications Interception and Access Act 1979 sitting over the top of eight separate state and territory surveillance and listening-device laws.
The federal law mainly targets third-party interception. The state laws govern the actual recording device - and they don't all read the same way.
This is why you'll find one article insisting Australia is "one-party consent" and another insisting it's "all-party consent."
Both are oversimplifying. Some jurisdictions clearly allow a participant to record; others frame recording of a "private conversation" as prohibited unless all parties consent or the recording is reasonably necessary to protect the recorder's lawful interests.
Definitions of "private conversation" and the available exceptions genuinely vary between, say, New South Wales and Queensland.
For a firm operating across multiple states, or taking calls from clients in states other than your own, trying to memorise the per-state position is a losing game.
The defensible approach is to treat disclosure as mandatory: tell every caller the line is recorded, give them a way to opt out, and don't rely on a state exception you haven't had confirmed. If you ever plan to use a recording as evidence or share it beyond the original parties, get specific legal advice first.
New Zealand: simpler, but the Privacy Act does the heavy lifting
New Zealand is a single-party (one-party) consent country. If you or your staff member are on the call, you can generally record it lawfully without the other party's explicit consent. The Crimes Act 1961 is aimed at intercepting communications you're not part of, not at recording your own.
That makes the "can I record?" question easy. The catch is the Privacy Act 2020, which governs everything that happens next. A recording that identifies someone is collecting personal information, which triggers obligations: collect it for a genuine purpose, be transparent about it, store it securely, and don't use or disclose it in ways a reasonable person would find objectionable. Publishing or sharing a recording outside its original purpose is where NZ firms most often get into trouble.
Practically: you can record without announcing it, but you generally shouldn't. A short recorded notice plus a line in your privacy policy covers the transparency expectation cleanly.
The UK: recording is easy, data protection is not
UK interception law (RIPA) explicitly lets a party to a call record it, and the Telecommunications (Lawful Business Practice) Regulations 2000 permit business recording for purposes like training, quality assurance and dispute resolution. So the recording itself is rarely the problem.
UK GDPR and the Data Protection Act 2018 are. A recording is personal data, so you need a documented lawful basis - for most firms that's legitimate interests, which the ICO expects you to back with a written Legitimate Interests Assessment. You must tell callers who's recording, why, and for how long; keep recordings only as long as needed; store them securely; and be able to answer a subject access request within one month. Note too that recent ICO guidance has pushed firms to be explicit when calls are analysed by AI — directly relevant if you use AI call summaries. Financial-services firms regulated by the FCA face stricter, longer retention duties again.
Retention: the part most firms get wrong
There's no universal "keep it for X months" rule. The principle in all three markets is the same: keep recordings only as long as you have a genuine reason to, then delete them on a defined schedule. Most businesses land somewhere between 6 and 24 months for general quality purposes - but a specific regulatory obligation (like Australia's tax-practitioner record-keeping rules, or FCA requirements in the UK) can override that and require years. Decide your retention period deliberately, write it down, and automate the deletion. "We just keep everything forever" is the answer that turns a data breach into a catastrophe.
A practical baseline that works in all three markets
If you do nothing else, consider doing these five things
- Announce that calls are recorded
- Give callers a way to opt out
- Write down your purpose and lawful basis
- Store recordings securely with restricted access
- Set a retention period. That baseline keeps you defensible whether your clients are in Auckland, Adelaide or Aberdeen.
This Cradle article is general information, not legal or professional advice. Record-keeping obligations vary by entity type, jurisdiction and circumstances, so check the current ATO or HMRC guidance and your professional body's requirements, and seek advice on your firm's specific situation.
