Why Business Emails Go to Spam (And How to Fix It)
By Jim Storey | Founder, Devout Digital | 9 min read
The quote went out on Tuesday morning. The price was sharp. The scope was clear. It was a job worth having. A week later, the buyer still hadn’t replied. You followed up and heard nothing. Later, you learnt that the work had gone elsewhere. Easy enough to blame a better price, an existing relationship, or timing you couldn’t control. There’s another possibility the sales review never checks: the buyer may never have seen your email.
Your Sent Folder Can't Tell You Where It Landed
The sent folder proves that your email left your mailbox. That’s all it proves. It can’t tell you whether the receiving service put it in the inbox, moved it to junk, or rejected it further down the line. When a system accepts a message and then filters it into junk, nothing bounces. Your team has nothing to investigate. As far as the recipient system is concerned, it delivered the email.
That’s what makes it easy to miss: there’s no evidence trail. A rejected address bounces, and somebody calls IT. An unseen quote just goes quiet. In sales, silence already has a dozen explanations, so nobody blames the email system. They blame the market instead.
The Riskiest Emails Are the Ones That Look Routine
Your most ordinary emails are often the most exposed ones. A tender response leaves at four o’clock against a five o’clock deadline. A quote goes to a buyer you’ve never contacted before. An order confirmation sits in a site manager’s junk folder while your team assumes the delivery went through. The receiving service doesn’t know one attachment represents weeks of work. It doesn’t know another message is holding up a purchase order. It just sees email, and scores whatever evidence comes attached.
That’s why this falls between sales and IT, and neither team owns it. It isn’t a campaign issue. Nobody unsubscribes from a quote. But the commercial damage still shows up, in conversion rates, missed deadlines, and buyers who seem to vanish. Good tender preparation still matters, as Maisona covers in its piece on winning government tenders before the RFT lands. But the finished response still has to survive the short trip from your system to the buyer’s inbox.
What Receiving Servers Check Before They Trust You
Email wasn’t originally designed to prove a sender is who they claim to be. So modern receiving systems rely on three authentication methods to build that confidence. The names sound technical. The questions behind them aren’t. Is this server allowed to send for your domain? Was this message signed by your domain? Does the authenticated identity match the address shown to the buyer?
SPF answers the first question. It’s a record attached to your domain that identifies the systems authorised to send email on its behalf. That means your main Microsoft 365 or Google Workspace service may pass, while a web server or quoting platform doesn’t. The record only helps when it reflects the systems you use.
DKIM answers the second question by adding a digital signature to each message. The receiving service checks that signature against a public key the signing domain publishes. A match shows that domain signed the message, and that nothing in it changed in transit. It’s a different route to trust, but it still doesn’t settle what the buyer sees in the From line.
DMARC connects the technical result to that visible identity. It checks whether a domain that passed SPF or DKIM aligns with the From domain. Then it lets the domain owner publish instructions for messages that fail. A third-party platform can pass SPF and DKIM for its own domain while failing DMARC for yours. That’s why “the software says authentication is enabled” isn’t always a useful answer.
The Systems You Trust Can Be the Ones That Fail
Your main mailbox is usually the easy part. The harder part is everything else that sends as your business, often without anyone thinking of it as part of the mail environment. Microsoft’s own guidance notes that organisations frequently publish incomplete SPF records. Usually, it’s because they haven’t identified every server, hosted application and software service using their domain.
A website form is a common example. If it sends from [email protected] through the web host, you need to authorise that host. The message needs its own aligned authentication path too. Owning both the site and the address doesn’t make the setup valid by itself. Without that configuration, a legitimate enquiry notification can resemble somebody impersonating your business.
Quoting software, an ERP and a CRM create the same gap in less visible places. Each service is a separate email sender. You need to authorise and configure each one to authenticate in line with your domain. Entering a salesperson’s address in the From field changes what the buyer sees, but it doesn’t change which infrastructure delivered the message.
Older systems make the inventory harder. An on-premises server can keep sending application mail after a Microsoft 365 migration, or an acquired business can keep using its former domain through the parent company’s systems. Neither arrangement is inherently wrong, but your authentication setup needs to account for each one. If you can’t name every sender, you can’t know whether a stricter policy will protect your mail or block it.
Gmail, Outlook and iCloud Are Already Enforcing This
Google made authentication a formal requirement for mail sent to personal Gmail accounts in February 2024. All senders need SPF or DKIM. Senders above 5,000 messages a day need both, plus DMARC, plus alignment between the visible From domain and at least one authenticated domain. Google counts that volume across the same primary domain, and once Google classifies a domain as a bulk sender, that classification doesn’t expire. From November 2025, Google began escalating enforcement on non-compliant traffic, with failures risking temporary and permanent rejections.
Yahoo introduced a similar distinction between all senders and bulk senders, although it deliberately doesn’t publish a numerical bulk threshold. Microsoft applies SPF, DKIM and DMARC requirements to domains sending more than 5,000 messages a day to Outlook.com consumer addresses; since 5 May 2025, high-volume messages that don’t meet those requirements can be rejected with an authentication error. Apple requires bulk senders to iCloud Mail to use SPF and DKIM, publish DMARC and meet its other sending requirements, but it also publishes no numerical threshold.
None of that means lower-volume business email gets a free pass. Google requires some authentication at every volume. Microsoft 365 uses SPF, DKIM and DMARC alongside other signals when it assesses incoming business mail. Authentication doesn’t guarantee inbox placement, but missing or misaligned authentication removes evidence that could have helped a legitimate message earn trust. What matters is whether you can test what you’re sending, whatever your volume.
You Can Test This Yourself in Minutes
Begin with a message from your normal work mailbox to a Gmail account you control. Open it in Gmail on the web, choose the three-dot menu and select Show original. Google displays separate SPF, DKIM and DMARC results. You want to see PASS on all three, with the authenticated domains matching or aligning with the business domain shown in the From address.
Then repeat the test from the systems that are easier to forget. Submit your website form, generate a quote from the quoting platform or ERP, and send a normal salesperson message from the CRM. A pass from your Microsoft 365 route proves only that route, so each system needs its own test.
You can also inspect what your domain publishes through Google’s free Admin Toolbox. Check MX reports configuration including SPF and DMARC, while the Dig tool inspects DNS records such as reverse DNS. These checks won’t repair anything. But they give you something far more useful than “our email seems fine”: a list of routes, results and visible gaps to take to your IT provider.
Give Your IT Provider the Full Picture
The request should be specific: “Please inventory every system that sends using our domains, confirm SPF, DKIM and DMARC alignment for each one, and set up DMARC aggregate reporting so we can see legitimate sources and failures.” Ask them to include the website host, Microsoft 365 or Google Workspace, CRM, quoting and ERP software, accounting platform, scanners, on-premises relays and acquired domains. You’re asking for a map of the sending environment. Not three green ticks against the main mailbox.
The sequence matters, because DMARC lets you tell receiving services to take no action, quarantine failures, or reject them outright. Starting at p=none lets you collect reports without asking receivers to block anything. That buys your provider time to find legitimate senders the inventory missed. The current DMARC specification recommends working through monitoring mode first. Then you tighten the policy in stages, once legitimate senders are passing.
Don’t set p=reject on day one. Say the website form, CRM or quoting platform is still failing when that policy takes effect. Receivers may then block exactly the mail you were trying to protect. Monitor first, correct every legitimate route, and tighten the policy in stages. The reports also need a named owner. An unread reporting mailbox is only a more technical version of the same silence.
Seven Things to Check Before You Call Your IT Provider
- Send a test from your normal mailbox and inspect SPF, DKIM and DMARC.
- Repeat the test from your website form, CRM, quoting platform and ERP.
- List every service and server that sends using your domain.
- Include acquired domains and old on-premises systems in that inventory.
- Ask your IT provider to confirm alignment, not merely that records exist.
- Begin DMARC in monitoring mode with aggregate reporting enabled.
- Move towards quarantine and rejection only after legitimate senders pass.
Enquiry volume healthy but conversion down? Or a tender response that went unanswered with no clear reason? Rule out an authentication gap before writing the loss off as price or market conditions. Your IT provider should own the technical repair. If the gaps run wider than one system, across the website, CRM and sales journey, talk to Maisona’s team about where those systems are letting opportunities disappear.
Frequently Asked Questions
Does this matter if you send fewer than 5,000 emails a day?
Yes. The 5,000-message figure triggers additional requirements at Google and Microsoft. But Google’s baseline requirements apply to every sender reaching personal Gmail accounts. A receiving service can also use authentication as one signal when evaluating a message. That’s true regardless of whether you meet a published bulk threshold.
Does Microsoft 365 handle this for you?
It can authenticate mail that leaves through Microsoft 365, but it can’t automatically cover unrelated infrastructure. Your website host, ERP, quoting platform and CRM may each need their own configuration. That’s why testing the main mailbox alone proves very little about the rest of your sending environment.
Will passing SPF, DKIM and DMARC keep every email out of spam?
No. Inbox providers also consider reputation, sending behaviour, recipient feedback and other abuse signals, so authentication isn’t an inbox guarantee. It establishes that your message has a credible claim to the identity it displays. That removes one avoidable reason for a legitimate quote to look suspicious.
Can you publish DMARC yourself?
The DNS record itself takes five minutes to write. Confirming what it should say takes longer. Before enforcement, somebody needs to identify every legitimate sender and configure alignment. Then they need to collect reports and confirm what a stricter policy would affect. A technically valid p=reject record can still cause commercial damage when the inventory behind it is incomplete.
How will you know which system is failing?
Individual test messages show the result for one route. DMARC aggregate reports show every source using your domain, and how receiving services evaluated them. Used together, the two turn a vague deliverability concern into something concrete: a defined list of systems your provider can investigate.
Jim Storey runs Devout Digital, a boutique Sydney web design and SEO agency. The agency helps small businesses across Sydney build clearer websites, improve their search visibility and generate more enquiries online.