1. Permission is the whole policy

Mail only people who asked to hear from you. Permission means the recipient took a deliberate action to opt in: filled in a form, ticked an unticked box, bought something from you and was told they would receive mail, or gave you their address in person for that purpose. You must be able to show when and how each contact opted in.

Permission is not transferable and does not survive:

• a purchase, rental, or trade of a list;
• scraping, harvesting, or guessing addresses;
• an acquisition, unless the contacts were told their data might transfer and the mail stays within what they opted into;
• a long silence. If a contact has not opened or clicked in two years, treat the permission as lapsed and re-confirm it or stop mailing them.

2. Lists you may not mail

• Purchased, rented, leased, traded, or otherwise third-party-sourced lists.
• Addresses collected by scraping websites, directories, social profiles, or WHOIS.
• Addresses generated by pattern or dictionary attack.
• Role accounts mailed at scale where no individual opted in, such as info@, sales@, or abuse@.
• Addresses on a suppression list, whether ours or yours, including anyone who has unsubscribed, complained, or hard-bounced.
• Any list you cannot describe the origin of. If you do not know where it came from, do not mail it.

3. Honest identification

Every commercial message must:

• use a From name and address that accurately identify the sender;
• carry a subject line that does not mislead about the content;
• include a valid physical postal address for the sender; and
• send from a domain you control and have verified with us.

Forged headers, misleading routing information, and impersonating another person or organization are prohibited outright. These are also legal requirements in most jurisdictions, including under the US CAN-SPAM Act, Canada's CASL, and the EU ePrivacy Directive, so the platform enforces several of them at render time rather than trusting the copy.

4. Unsubscribes

Every commercial message must carry a working, one-click unsubscribe that requires no login, no password, and no reason. The platform injects one automatically where your template does not already have one, and you may not remove, obscure, or disable it.

Honour an unsubscribe immediately, and in no case later than ten days. Once someone unsubscribes you may not mail them again on that list without a fresh opt-in, and you may not move them to another list to keep mailing them.

Transactional messages, meaning mail sent to complete a transaction the recipient initiated, are exempt from the unsubscribe requirement. Marketing content bolted onto a transactional message makes the whole message marketing.

5. Prohibited content

You may not use the Service to send, host, or link to:

• phishing, credential harvesting, or anything designed to deceive a recipient into disclosing information;
• malware, ransomware, or malicious code, including in attachments and linked files;
• content that infringes intellectual property or misappropriates a trade secret;
• child sexual abuse material, which we report to the authorities without notice to you;
• content that harasses, threatens, defames, or incites violence, or that promotes hatred against people on the basis of a protected characteristic;
• unlawful gambling, controlled substances, weapons, counterfeit goods, or forged documents;
• pyramid schemes, chain letters, "get rich quick" offers, and comparable financial deception;
• anything otherwise unlawful in the jurisdiction of the sender or the recipient.

Lawful adult content, and content on subjects a general-audience platform might find awkward, is permitted where the audience genuinely opted in, the sender complies with applicable law, and the content is not otherwise prohibited above. The standard here is legality and consent, not taste.

6. Prohibited technical use

• Probing, scanning, or testing the security of the Service, or circumventing any authentication, isolation, rate limit, or resource limit.
• Accessing another customer's data, or attempting to.
• Using the Service as an email-verification or list-cleaning oracle, meaning uploading addresses primarily to harvest bounce feedback rather than to communicate with them.
• Creating multiple accounts to evade a limit, a suspension, or a plan allowance.
• Automated load that degrades the Service for others, including deliberately abusive imports. See the Service Limits.
• Reselling or white-labelling the Service without a written agreement with us.

7. The thresholds we enforce

These are not guidelines. The platform measures them continuously per account over a rolling seven-day window and acts automatically when one is breached with enough volume for the rate to be meaningful.

Rates are only evaluated once an account has attempted at least 50 messages in the window, below which a single bounce is statistical noise rather than a signal.

8. How we enforce

A breach of a threshold re-sandboxes the account. External sending stops immediately. Your data, your templates, your reports, and internal test sending to your own team all keep working, and you can export everything. We notify the account owner by email with the measured rates.

Lifting a suspension is a human decision, never automatic. Write to support@sentientmail.com with what happened and what you changed. Cleaning the list, re-confirming permission, and reducing send volume are the usual answers.

For breaches of sections 1 through 6 we may warn, suspend sending, suspend the account, or terminate it, depending on severity and on whether the harm is ongoing. Phishing, malware, and illegal content are terminated on confirmation, without warning. Where we can, we warn first, because most breaches are mistakes.

9. Reporting abuse

If you received mail through SentientMail that you did not ask for, report it to abuse@sentientmail.com. That inbox is monitored continuously and reports are actioned ahead of other mail.

Forward the message with full headers if you can. Headers identify the sending account, which is what lets us act on it. Without them we often cannot tell which of our customers sent the message.

Using the unsubscribe link in the message is usually faster than writing to us, because it stops that sender at once. Report it to us when the sender ignores the unsubscribe, when you never opted in at all, or when the mail looks like phishing.