How to check if incoming emails go to spam

Spam filtering happens after a message has already been accepted. The receiving server takes the mail, then decides where to put it — and that decision is invisible to the sender, who sees only a successful delivery. This is why "I definitely sent it" and "it never arrived" can both be true.

This covers where filtered mail actually goes, how the decision is made, and what genuinely changes the outcome.

Where filtered mail ends up

Not everything goes to a folder called Spam, and this is where most searches fail.

Gmail files mail into Spam, but also sorts into Promotions, Social, and Updates tabs, which many people never look at. Search in:spam or, more usefully, in:anywhere — the latter includes Spam and Trash, which a normal search excludes.

Outlook and Microsoft 365 use Junk Email, plus a Focused/Other split that hides low-priority mail in a second tab. Organisational accounts add a further layer: a hosted quarantine that holds mail before it reaches the mailbox at all. Users cannot see it; an administrator has to review and release.

Yahoo and AOL use a Spam folder with a retention period, after which messages are deleted permanently.

Apple Mail uses Junk, and applies filtering both at the iCloud server and again locally in the app — so a message can be filtered on one device and not another.

Google Workspace adds admin-level quarantine on top of the user's Spam label, configured by policy.

The practical instruction: search every folder including Trash and Archive, and on a work account, ask an administrator to check quarantine. A message absent from every user-visible folder is frequently sitting in quarantine.

How the decision is made

Filtering is a scoring process combining several independent signals. No single one determines the outcome.

Authentication

The strongest signal available. SPF, DKIM, and DMARC let the receiver verify the sender is authorised to use the domain it claims. Failures push a message toward spam and, under a strict DMARC policy, cause outright rejection.

This tightened substantially in 2024, when Google and Yahoo began requiring SPF and DKIM on all bulk mail, DMARC for high-volume senders, one-click unsubscribe headers, and spam complaint rates kept below a low threshold. Microsoft introduced comparable requirements for consumer Outlook accounts afterwards. Senders who had been fine for a decade suddenly found their mail filtered. Google publishes the current sender guidelines in full.

Sender reputation

Providers track how recipients treat mail from an IP address and a domain over time. Complaints, deletions without reading, and messages to nonexistent addresses all lower reputation. Opens, replies, and rescues from spam raise it.

Two situations produce poor reputation through no fault of the sender: a shared hosting IP where another customer behaves badly, and a brand-new IP or domain with no history — absence of reputation is itself treated with suspicion.

Recipient behaviour

Your own history matters more than most people realise. If you have previously replied to a sender, or added them to contacts, their mail is far more likely to reach the inbox. If you or others on the same provider have marked similar mail as spam, the opposite applies.

Content and structure

Weaker than commonly believed, but real. Large image-to-text ratios, link shorteners, mismatched display names and actual addresses, and certain attachment types raise the score. Individual "trigger words" are largely a myth at this point — modern filters weigh patterns, not vocabulary lists.

Infrastructure signals

Missing or mismatched reverse DNS on the sending server, a HELO name that does not match, or a return path that does not align with the visible sender all contribute.

What actually changes the outcome

Ranked roughly by effect.

Reply to the sender, or add them to contacts. The single most effective recipient-side action. A reply is a strong positive signal on most platforms.

Mark messages as "Not spam" rather than dragging them out. Using the explicit control sends feedback to the filter. Dragging a message to the inbox often does not.

Create a filter that never sends a specific sender to spam. Gmail supports this directly through filter settings; Outlook's Safe Senders list serves the same purpose. This is more reliable than contact-based allowlisting.

Ask an administrator to allowlist at the gateway. On business accounts, gateway policy overrides anything the user does in their mailbox.

Fix authentication if you control the sending domain. Where the sender is your own system — a website form, an application, a CRM — correct SPF, DKIM, and DMARC records typically resolve the problem outright.

Stop marking legitimate mail as spam. Filters learn from corrections in both directions, and a mistaken report teaches the wrong lesson.

What does not help: asking senders to remove words from their subject lines, or repeatedly resending the same message.

Confirming what is happening

Because acceptance and placement are separate, a useful check observes both.

Send a message from an independent source and compare the SMTP result against where it lands. If delivery status reports acceptance and the message is in Spam, filtering is the whole story and nothing is wrong with your routing. If it reports acceptance and the message is nowhere, look at rules, forwarding, and quarantine. If it is rejected outright, the problem is not filtering at all — see why you're not receiving emails.

A test message landing in Spam is also, in itself, normal. An unfamiliar sender with no prior relationship to your mailbox is exactly the profile filters treat cautiously.

Frequently asked questions

Why does one sender always go to spam when everyone else is fine?

Almost always authentication or reputation on their side. Ask whether they have SPF and DKIM configured — if they send from a website form or an application, often nobody ever set them up.

I marked it "not spam" and it happened again. Why?

Feedback influences the filter but does not override it, especially when authentication is failing. A filter rule or Safe Senders entry is more decisive than repeated corrections.

Can I stop spam filtering entirely?

On most consumer accounts, no. Administrators of business accounts can weaken policy, but disabling filtering exposes the mailbox to the full volume of unsolicited mail, which is rarely a good trade.

Does opening a spam message make things worse?

Not for filtering. It can confirm the address is live if the message loads remote images, which is a reason to leave image loading off by default.

How long do messages stay in Spam?

Typically 30 days, after which most providers delete them permanently. Anything you need should be moved out well before that.

Is a message in Spam the same as a message that was blocked?

No. Spam means the server accepted it and filed it. Blocked means it was refused at SMTP and never arrived — the sender receives a bounce in that case.

---

To see whether a mailbox is filtering rather than failing, send a test message and compare the delivery status with where it lands.