Two kinds of regime
Anti-spam laws split into opt-out regimes, where you may email first and must stop when asked, and consent regimes, where you need permission (express or implied) before the first email. The United States is opt-out. Canada and Australia require consent. The UK and EU sit between, depending on whether the recipient is a company or a person and on the GDPR status of their data.
Where you are based matters less than where your recipient is. A US company emailing Canadian prospects is subject to CASL. Relevance comes up again and again: France, Canada and Australia all tie permissibility to whether the message relates to the recipient's professional role [3,5,9].
United States: CAN-SPAM
The FTC is explicit that “the law makes no exception for business-to-business email” [1]. CAN-SPAM requires no prior consent, but every commercial email must:
- Use accurate header and routing information (From, To, Reply-To).
- Use a subject line that isn't deceptive. A fake “Re:” on a first email fails this.
- Identify the message as an advertisement where applicable.
- Include a valid physical postal address: a street address, a registered P.O. box or a commercial mail receiving agency.
- Explain clearly how to opt out, with a mechanism that works for at least 30 days after sending.
- Honor opt-outs within 10 business days, without charging a fee or asking for more than an email address.
- Remain your responsibility even if an agency or tool sends for you.
- Penalties run up to $53,088 per violating email as of the January 2024 inflation adjustment [1]; the figure is adjusted yearly.
United Kingdom: PECR and UK GDPR
Under the UK's PECR rules you may email corporate subscribers (companies, LLPs, government bodies) without consent, provided you identify yourself, give a valid address for opt-outs and keep a do-not-email list [2]. Individual subscribers, including sole traders and some partnerships, need consent or the “soft opt-in”, which covers only existing customers for similar products and never bought-in lists [2].
PECR not requiring consent doesn't mean anything goes. A named employee's address is personal data, so UK GDPR still needs a lawful basis, usually legitimate interests [2]. The ICO's updated guidance (March 2026) says legitimate interests can support direct marketing where it is proportionate and unsurprising, and the right to object to direct marketing is absolute [4].
The Data (Use and Access) Act 2025 raised the maximum PECR fine from £500,000 to £17.5 million or 4% of global turnover [6]. The government considered extending consent rules to B2B email and decided not to [6].
European Union: ePrivacy and GDPR
Article 13 of the ePrivacy Directive requires prior consent for unsolicited email marketing to natural persons, and leaves member states to protect legal persons' interests as they choose [7]. B2B rules therefore vary by country.
France is a useful example: the CNIL allows email to professionals without prior consent if the message relates to their job, such as software pitched to an IT director, and they can object easily. Generic addresses like info@ fall outside these principles [3]. Check each target country's rule before sending; don't assume France's applies everywhere.
GDPR applies to named business contacts everywhere in the EU. The EDPB's guidance on legitimate interest stresses that direct marketing doesn't automatically qualify: the purpose, necessity and balancing test still applies [8]. In practice: document a legitimate interest assessment, keep messages role-relevant, tell people in the first email where you got their data and how to object, and honor objections permanently.
Canada: CASL
CASL is the strictest of the five for cold outreach. You need consent before sending a commercial message, plus sender identification and an unsubscribe mechanism [5].
Implied consent exists in narrow cases: an existing business relationship (a purchase or contract in the past two years, or an inquiry in the past six months), or “conspicuous publication”, where the recipient published the address themselves, didn't say “no unsolicited messages”, and your message is relevant to their role [5]. The burden of proving it falls on you, and addresses from third-party lists generally won't qualify [5].
Unsubscribes must be processed within 10 business days and the mechanism must work for at least 60 days after sending. Penalties reach C$10 million per violation for organizations [10].
Australia: the Spam Act
Australia requires consent, express or inferred, plus accurate sender identification and a working unsubscribe [9]. Inferred consent can come from a conspicuously published address, but only if the message relates to the recipient's role [9].
Unsubscribes must be actioned within five working days, cost nothing, work for at least 30 days, and require no login or extra personal information [11]. The regulator enforces this: ACMA penalized Telstra A$626,000 in March 2025 because its unsubscribe process asked for a PIN or extra personal details [12].