Could a scammer email your customers as you? Lock down your newsletter account

When someone gets into your newsletter tool, the scam goes out under your name, from your domain, and passes every check. Here's what two incidents this year showed, the three dull settings I'd change today, and what to say to customers if it happens.

Patterned paper envelopes, stationery and a half-finished piece of knitting on a weathered wooden table in soft daylight

On 10 September 2026 an email marketing platform published an incident write-up that anyone who sends a newsletter should read once. An attacker had got into 138 customer accounts. From 43 of them the contact lists were exported, and 6 were used to send phishing to the people on those lists. The platform's own description of those emails: they "passed the usual email authentication checks and looked genuine".

They looked genuine because, in the dull technical sense, they were. Right sender, right domain, right logo, sent from the same account that sends the real newsletter every month. Only the link was wrong.

I'm not naming the platform or the companies. Which tool it was doesn't change much for you, because the same thing can happen through any of them, and the fixes below are the same whichever one you use.

What happened, in plain words

Two incidents from this year show the two usual ways in.

In September the hole was on the platform's side. According to its write-up, the attacker created their own account with single sign-on, invited real users into it, and a boundary that should have kept that login inside one company's account didn't hold, so the attacker could reach every account those users could reach. The platform closed the route two hours after spotting it and signed every user out. One of the affected companies said it took the scam link's domain down within 20 minutes. It could, because the platform routed that company's mail through the company's own domain.

In April a writer with a large newsletter published a notice explaining that someone had got into their web server and used it to reach the connection between the website and the email platform. Three fraudulent emails went out to the mailing list; a fourth was blocked. The clean-up list in that notice is worth copying. They removed the attacker's access, revoked and regenerated every API key, forced password resets, re-secured the email integration and reported the incident to the UK's data regulator.

So in one case the email tool itself let people down, and in the other the website was the door. You can't do much about the first except notice fast and react well. The second is mostly in your hands.

Why "check the sender" doesn't help here

Most anti-phishing advice tells customers to look at the sender address. That works against forgeries. It does nothing when the message really comes from you.

SPF, DKIM and DMARC are the three records that let inboxes check whether a message came from a server you allowed to send for your domain. Since 1 February 2024 Google has required all three from anyone sending more than 5,000 messages a day to Gmail accounts (Google's sender guidelines), so most newsletter senders already have them. They prove the mail left a server you approved for your domain, and they say nothing at all about who sat at the keyboard, so if someone is logged into your account or holding your API key, the mail they send is signed exactly like yours.

That's why the protection has to sit one step earlier, at the account.

The three boring fixes

None of this is clever, and that's exactly why it gets skipped.

Two-step login (or SSO) on the email tool, for everyone. Your newsletter account can write to every customer you have, so treat it like your bank login rather than a social media scheduler. Turn on 2-step verification for every user, not only the owner. The UK's NCSC guidance on 2-step verification prefers an authenticator app to text messages and says any second step beats none. If your company already signs in to everything through one identity provider, connect the email tool to it, so people who leave lose access on the day they leave. There's a limit I should be straight about. In the September case the weak point was the platform's own single sign-on handling, and no setting on the customer's side would have stopped it. What I don't know, and the write-up doesn't say, is whether any of the six sending accounts had extra protection that slowed the attacker down. I'd still switch 2FA on everywhere, because a reused password is the far more common way in.

Only one or two people can export the list. Open the user list in your email tool and read it slowly. Former employees. The freelancer who ran one campaign in 2023. A test login nobody remembers making. Most tools have roles, and someone who writes campaigns rarely needs to download every contact. Keep export and admin rights for one or two named people, and remove everyone else or drop them to the lowest role that still lets them do their job.

Delete or rotate API keys you don't use. An API key is a password that never asks for a second step. In the April case the attacker reached the email tool through the website's connection to it, and connections like that run on keys. Open the list of keys in your email tool. If you can't say what one is for, switch it off and wait to see whether anything complains; if the signup form on your site stops working, you've found its key and can issue a new, narrower one. Where the tool lets you limit what a key can do, give the website's key the smallest job it needs (for a signup form, adding a subscriber) and never full sending rights.

Watch for look-alike domains

Even with your account locked, someone can register a domain that looks like yours with two letters swapped and send from that. Your own SPF, DKIM and DMARC records protect your real domain, not the copy. A few habits make the fake easier to spot. Send from one domain, always, and say so on your site. Turn on DMARC reports, which Google's own guidelines recommend, so you can see who is sending mail in your name. Now and then, search for your business name with obvious typos. If you find a copy, report it to the registrar and the host. And while you're in the registrar account, here's how to protect the real domain from hijacking.

What to tell customers if it happens

Keep it short and send it fast. The April notice is a decent template, and it does five things: it says what happened and when, which data was involved, what was done about it, what to do if you clicked, and what a real email from them will never ask for.

Send it through a channel you know is clean, put the same text on your website, and leave every link out of the warning except one to that page. If personal data left your hands, a legal clock may be running too: Article 33 of the GDPR asks you to tell your data protection authority within 72 hours of becoming aware, unless the breach is unlikely to put people at risk, and the UK version has the same 72 hours. The details depend on your case and your country, so get proper advice. I'm not your lawyer.

Where your website forms send data

Form connections are the part I look at first on most sites, because people forget they exist. The newsletter signup box, the contact form and the checkout each hand data to other tools: the email platform, a CRM, a spreadsheet somebody set up years ago. Every one of those connections runs on a key or a login stored somewhere on your web server, and if that server runs outdated software, whoever gets in can pick the key up and use it, which is the April story almost word for word. That's why I count the plugin and PHP updates from the website security check as newsletter security too.

Keep those keys on the server, never in the page's own code, and make sure the form can only add a subscriber, never send a campaign. Updates and keeping track of keys are the kind of work website maintenance should cover. When we build a site, the form connections are written down: which tool, which key, what it can do, who can change it.

If you'd like a second look at how your forms and email tool are wired together, get in touch or write to contact@magnuszgr.com with your site address and the name of the tool you use. No passwords in the first email, please.

Your newsletter is the one channel where customers already trust the sender. Guard the account like it's worth exactly that.

Read next