Domain health guide
Gmail, Yahoo and Outlook sender requirements, in plain terms
The three largest consumer mailbox providers now require authenticated mail from bulk senders. The rules are similar but not identical, and several apply to every sender regardless of volume.
The timeline
Google and Yahoo began enforcing new sender requirements in February 2024. Microsoft announced equivalent requirements for Outlook.com, Hotmail.com and Live.com addresses, in effect from May 2025. In each case enforcement was phased in: first warnings and spam-folder placement, later temporary and permanent rejections.
Who counts as a bulk sender
Google defines a bulk sender as a domain that sends around five thousand or more messages a day to personal Gmail accounts. Microsoft uses the same five-thousand threshold for its consumer domains. Messages are counted by the domain in the visible From address, so several systems sending as the same domain are added together.
Smaller senders are not exempt from authentication. Google's guidelines ask every sender for SPF or DKIM, valid forward and reverse DNS for sending IPs, and TLS for transmission.
What bulk senders must have
The common core is the same across the three providers.
- SPF and DKIM both set up for the sending domain.
- A DMARC record published for the domain. A policy of p=none is accepted; it does not have to be quarantine or reject.
- Alignment: the domain in the visible From address must match the SPF domain or the DKIM signing domain.
- A working one-click unsubscribe in marketing and subscribed mail, honoured within two days (Google and Yahoo).
- A low spam complaint rate. Google asks senders to stay below 0.3 percent as reported in Postmaster Tools.
What p=none does and does not give you
Publishing p=none satisfies the DMARC requirement, which is why many domains stop there. It does not protect the domain: receivers are asked to deliver failing mail as usual.
The value of p=none is the reports. With a rua address, receivers send aggregate reports that list every source sending mail as your domain and whether it passed. Those reports are what make it possible to move to quarantine or reject without losing legitimate mail.
How to check a domain
Read the published records rather than the intended configuration. Confirm there is exactly one SPF record and that it stays within ten DNS lookups, that the selectors your senders use publish valid DKIM keys, and that a single DMARC record exists with a report address.
Then confirm alignment using real mail: send a message to a mailbox at each provider and read the Authentication-Results header. Records can be correct while alignment still fails because a third-party service signs with its own domain.
What this means for agencies and MSPs
If you manage domains for clients, one misconfigured client domain produces a deliverability incident that the client experiences as their email being broken. Checking each domain once is not enough, because records change when a client adds a tool or a provider changes its includes. Continuous monitoring with change alerts is what turns this from an incident into a routine fix.
The minimum set of records
For a domain that sends through one provider, three records cover the authentication part of the requirements. They are shown here for a domain using a mailbox provider and signing with its own domain.
example.com. TXT "v=spf1 include:_spf.google.com ~all"
s1._domainkey.example.com. TXT "v=DKIM1; k=rsa; p=MIIBIjANBgkqh..."
_dmarc.example.com. TXT "v=DMARC1; p=none; rua=mailto:dmarc@example.com"One-click unsubscribe, concretely
The unsubscribe requirement is technical, not just a link in the footer. Marketing and subscribed messages must carry two headers defined in RFC 8058. The first gives an HTTPS address that accepts a POST request; the second declares that one request is enough. Mailbox providers use them to show their own unsubscribe button, and the recipient must be removed without having to log in or confirm on a further page.
A visible unsubscribe link in the message body is still required. Transactional messages such as receipts and password resets are not covered by this rule. Most sending platforms add the headers for you; mail generated by your own application usually needs them added by a developer.
List-Unsubscribe: <https://example.com/unsubscribe?id=abc123>
List-Unsubscribe-Post: List-Unsubscribe=One-ClickReverse DNS and TLS
Two requirements apply to the sending server rather than to the domain. Each sending IP address must have a PTR record, and the hostname in that record must resolve back to the same address. This is called forward-confirmed reverse DNS. Mail must also be transmitted over a TLS connection.
If you send through a mailbox provider or an email service, they manage both and there is nothing for you to configure. If you run your own mail server, the PTR record is set by whoever owns the IP address, normally the hosting company, not in your own DNS zone.
dig -x 203.0.113.10 +short ; mail.example.com.
dig mail.example.com A +short ; 203.0.113.10Complaint rate
The complaint rate is the share of delivered messages that recipients mark as spam. Google asks senders to keep it below 0.1 percent and never let it reach 0.3 percent. It is calculated by the mailbox provider and shown only in its own tools, which you can access after verifying your domain there.
It is not something DNS can fix. The rate goes down when people receive only mail they asked for, when unsubscribing is easy and takes effect quickly, and when addresses that never engage are removed. A purchased or very old list is the usual cause of a high rate.
What a rejection looks like
When a message is deferred or refused for one of these reasons, the receiving server says so in its SMTP response, and the text appears in the bounce message or in your sending platform's logs. The response names the missing element, such as authentication, alignment or reverse DNS, and usually links to the provider's guidance.
Read the response rather than guessing. A temporary response, with a code beginning with 4, means the message will be retried and is a warning. A permanent response, beginning with 5, means the message was refused. The enhanced status code that follows often starts with 5.7, the family used for policy and security refusals.
A checklist to run for each sending domain
The requirements translate into a short list that can be checked for every domain and every system that sends as it.
- Exactly one SPF record, valid, within ten DNS lookups.
- DKIM signing enabled at every sending system, with a key of at least 1024 bits and preferably 2048.
- A DMARC record with at least p=none and a rua address.
- The From domain aligned with the SPF domain or the DKIM signing domain for every sender.
- One-click unsubscribe headers on marketing and subscribed mail, and requests honoured within two days.
- Valid forward and reverse DNS for sending IP addresses, and TLS for transmission.
- Complaint rate checked in the mailbox providers' own tools.
- A From address at your own domain, never at a free mailbox provider's domain.
Do not send as a free mailbox address
Sending bulk mail through your own platform with a From address at a consumer mailbox domain cannot be made compliant. Those providers publish DMARC policies of quarantine or reject for their own domains, and you cannot sign for or align with a domain you do not control, so the messages fail DMARC and are filtered or refused. Use an address at a domain you own.
Smaller senders
The five-thousand threshold decides who must meet the full set, but it is not a safe harbour below it. All senders are expected to authenticate with SPF or DKIM, and unauthenticated mail is increasingly deferred or filtered whatever the volume. The count is also per domain and cumulative across systems, so a small business whose newsletter platform, shop and staff all send as one domain can cross the threshold on a busy day. Meeting the bulk-sender requirements costs little and removes the question.
Staying compliant
Compliance is a state that decays. A new tool is connected and sends unaligned mail. A provider changes the records behind an include and SPF goes over the limit. A DKIM key is rotated at the provider and the DNS record is not updated. Each of these breaks a requirement without anyone editing the DMARC record. Checking once at setup shows that the domain was compliant on that day; monitoring the records and reading the aggregate reports is what shows that it still is. CloudSpex checks the public records on a schedule and alerts when one changes; it does not measure complaint rates or inbox placement.
Frequently asked questions
Is DMARC p=none enough for Gmail and Yahoo?
Yes, for the requirement itself. A DMARC record with p=none is accepted. It does not ask receivers to block failing mail, so it is a starting point, not protection.
Do the rules apply to senders under 5,000 messages a day?
The bulk-sender rules apply above the threshold, but every sender is asked to authenticate with SPF or DKIM, and unauthenticated mail is more likely to be deferred or filtered.
Does CloudSpex measure my spam complaint rate?
No. Complaint rates are only visible in the providers' own tools, such as Google Postmaster Tools. CloudSpex checks the public DNS records and reports.
Do transactional emails need an unsubscribe header?
No. The one-click unsubscribe requirement applies to marketing and subscribed messages. Receipts, password resets and similar messages are exempt.
Does the 5,000 threshold count per sending system?
No. It counts messages by the domain in the From address, so all systems sending as the same domain are added together.
Will p=none be enough in the future?
It satisfies the requirement as currently published. Providers have signalled that expectations will rise over time, and moving to an enforcing policy is the direction to plan for.
Use public checks as a starting point
CloudSpex reads limited public-facing signals. It does not change DNS, website, certificate, or email settings. Confirm the responsible provider and test relevant changes before applying a remediation.
Editorial details
Author and reviewer: CloudSpex. Sources: public technical standards and provider documentation relevant to the topic. Last reviewed: November 10, 2026. We update guides when the product behavior or the underlying guidance changes.