SPF, DKIM and DMARC: stop fraudsters sending email as your firm
Fraudsters can send email that appears to come from your firm's own domain. SPF, DKIM and DMARC make that much harder. Here is what each one does and how to introduce them without disruption.
Email was designed at a time when everyone on the network was assumed to be trustworthy. By default, nothing stops a stranger from sending a message that claims to come from your firm’s domain. The sender’s name and address shown in the recipient’s inbox can simply be typed in.
For most businesses that is a nuisance. For a wealth manager or family office it is a genuine risk. Your clients, their lawyers and your custodians act on instructions that arrive by email, and a convincing message from your domain is exactly what a fraudster wants.
Three DNS records, known as SPF, DKIM and DMARC, make spoofing your exact domain much harder. None is new or exotic, but they are often half-configured, and the missing half is usually the one that matters.
Why wealth firms are attractive targets
Email fraud relies on plausibility rather than clever technology. A message that appears to come from a relationship manager, asking a client to confirm new bank details, or from a director, asking a colleague to release a payment before the end of the day, can succeed because it fits the normal rhythm of the business.
Wealth firms combine several things that make this attractive:
- regular movements of significant sums
- clients who are used to dealing with you by email
- small teams, where one person may be able to act quickly
- relationships with lawyers, accountants and banks that also communicate by email
Authentication records do not stop every kind of email fraud, but they remove the easiest version of it: a message that carries your own domain in the sender’s address.
SPF: who is allowed to send
SPF, the Sender Policy Framework, is a single text record in your domain’s DNS. It lists the services allowed to send email on behalf of your domain, such as your email provider and any other systems that send in your name.
When a receiving mail server gets a message, it can check whether the server that delivered it is on your list. If it is not, the message fails SPF.
SPF has limitations worth knowing:
- it checks the technical “envelope” sender, which the recipient does not see, rather than the address shown in the inbox
- it can break when email is forwarded
- a record is limited to ten DNS lookups, so a long list of included services can quietly stop working
SPF is useful, but on its own it does not protect the address your clients actually see.
DKIM: a signature on every message
DKIM, DomainKeys Identified Mail, adds a digital signature to each outgoing message. Your sending service holds a private key and signs the message; the matching public key is published in your DNS. The receiving server uses that public key to confirm that the message was signed by an authorised sender and has not been altered on the way.
Each service that sends in your name should sign with its own DKIM key on your domain. Keys should be of adequate length and replaced from time to time, and old keys removed when a service is no longer used.
DMARC: the instruction that ties them together
DMARC, Domain-based Message Authentication, Reporting and Conformance, is the record that makes SPF and DKIM meaningful for the address your recipients see. It does two things.
First, it requires alignment: the domain that passes SPF or DKIM must match the domain in the visible “From” address. This is what stops a fraudster from passing SPF on their own domain while displaying yours.
Second, it tells receiving servers what to do with messages that fail, through a policy:
- p=none asks for monitoring only; failing messages are delivered as usual
- p=quarantine asks receivers to treat failing messages as suspicious, usually by sending them to spam
- p=reject asks receivers to refuse failing messages outright
DMARC also lets you request reports. Receiving providers send aggregate reports showing which servers sent email using your domain and whether those messages passed. These reports are the key to moving forward safely.
Moving from monitoring to enforcement
Many firms publish a DMARC record with p=none and stop there. That is a reasonable first step, but it does not protect anyone: spoofed messages are still delivered. The aim is to reach quarantine and, where possible, reject.
Doing that safely is a matter of patience rather than complexity:
- Publish DMARC at p=none with reporting enabled.
- Review the reports over several weeks to identify every legitimate source of email using your domain.
- Correct SPF and DKIM for each legitimate source until its messages pass and align.
- Move to p=quarantine, optionally applying it to a percentage of failing messages at first.
- Continue reviewing reports, then move to p=reject once you are confident nothing legitimate is failing.
Rushing to reject before step three is complete risks blocking your own invoices, newsletters or client portal notifications.
Do not forget your other senders
The services that most often fail are not your main mailbox. They are the systems added over the years, each sending email in your firm’s name:
- your CRM and client portal
- newsletter and event tools
- accounting, invoicing and e-signature services
- website contact forms and booking systems
- recruitment platforms
Each one needs to be included in SPF where appropriate, set up to sign with DKIM on your domain, or moved to send from a dedicated subdomain. Domains and subdomains that never send email should say so explicitly, with an SPF record that authorises nothing and a DMARC policy of reject.
The requirements Google and Yahoo introduced for bulk senders in February 2024 made this more pressing: organisations sending large volumes to their users must authenticate with SPF and DKIM and publish a DMARC record, among other conditions. Even at modest volumes, a newsletter tool sending in your name benefits from the same discipline.
BIMI: optional, and only after enforcement
BIMI, Brand Indicators for Message Identification, lets supporting email providers show your logo beside authenticated messages. It requires DMARC at quarantine or reject, and some providers also require a certificate for the logo. It is a pleasant finishing touch, but adds no protection beyond the enforcement it depends on.
What these records do not do
Authentication protects your exact domain. It does not stop a fraudster from registering a similar-looking domain, or from sending from a free email account with your partner’s name as the display name. Nor does it help if a genuine mailbox is compromised.
Above all, it does not replace the procedures that already protect your clients. Changes to payment instructions should continue to be verified by calling back on a number already held on file, never one supplied in the message itself. Clients should know that your firm will never change bank details by email alone. Your compliance officer will have views on how these procedures are documented and communicated; the records described here support them but cannot stand in for them.
A practical checklist
- Confirm SPF exists, lists every legitimate sender and stays within the lookup limit.
- Confirm each sending service signs with DKIM on your domain.
- Publish DMARC with reporting, and have someone actually read the reports.
- List every third-party service that sends in your firm’s name.
- Protect domains that never send email with restrictive SPF and DMARC records.
- Move gradually from p=none to quarantine, then to reject.
- Keep call-back verification for any change to payment instructions.
Email authentication is not a one-off project. Services are added, keys need rotating and reports need reading. It is part of what we look after in digital presence care, alongside the website and the profiles that carry your firm’s name.