Why Are My Emails Going to Spam? Causes and Fixes

If your emails are landing in spam, the fastest answer is rarely a new subject line. First determine whether the receiving server rejected the message, accepted it and filed it in spam, or filtered it only for one provider or recipient group. Then inspect the message headers and sender signals before changing the campaign.

This guide groups practical diagnostic categories, not frequency-ranked causes: authentication and alignment problems, list quality and complaints, sending-pattern or infrastructure changes, misleading message elements, and recipient-side policies. It also shows what Gmail and Outlook evidence can and cannot tell you.

Start with a copy of one message that landed in spam and, if possible, one recent message from the same stream that reached the inbox. Keep the recipient provider, send time, message ID, campaign, and sending service beside both samples. That small evidence set prevents a content guess from replacing the actual diagnosis.

Keep message headers and recipient samples in approved private systems. Redact email addresses, message IDs, IP addresses, tokens, and other identifiers before sharing anything outside those systems.

Step 1: Identify the symptom

The symptom determines which evidence to collect first. A bounce is not the same problem as a message that was accepted and moved to Junk.

Symptom	What it points you toward	First check
A bounce or SMTP error	A server-side authentication, DNS, policy, or rate issue	Read the complete code and enhanced status text; do not troubleshoot from the subject line
Spam at several providers	A sender-wide authentication, reputation, list, or infrastructure issue	Compare headers from two providers and review recent changes
Spam at one provider only	A provider-specific reputation, policy, or recipient rule	Use that provider’s headers, dashboard, or quarantine evidence
A sudden change	A DNS edit, key rotation, new sending service, list import, volume change, or policy change	Compare the last known-good message with the first failed one
One campaign or segment is affected	Links, content, audience expectations, or that segment’s list quality	Send a controlled comparison and inspect bounces and complaints
Messages arrive but replies or clicks fall	Filtering before recipients can act, or a change in audience intent	Use placement evidence and provider data; do not treat a low visible complaint count as proof of inbox delivery

Step 2: Read the evidence in the message

Inspect authentication results

In Gmail, open the message and choose More > Show original. Google’s header guide explains how to copy and analyze the full header. Look for the Authentication-Results line and record the result and domain for each method:

SPF checks whether the sending server is authorized for the envelope or Return-Path domain.
DKIM checks the cryptographic signature and the signing domain shown in d=.
DMARC checks whether SPF or DKIM passes with alignment to the visible From: domain.

A pass from an unrelated service domain does not necessarily give the visible From domain DMARC alignment. Check every platform that sends as your domain, including transactional, marketing, support, and outbound tools. If a new tool sends with a different Return-Path or DKIM signer, that change can explain a sudden drop even when the main provider still looks healthy.

Read provider-specific headers

For Microsoft 365, find X-Forefront-Antispam-Report in the message details. Microsoft’s current header reference documents useful fields such as CAT, SFV, DIR, PTR, and SRV. SFV:SPM means spam filtering marked the message as spam; CAT identifies the policy category that acted. In cloud organizations, Microsoft says SCL does not determine the final cloud verdict, so use CAT and DIR rather than treating SCL alone as the answer.

If a recipient cannot share headers, ask whether the message was placed in Junk, quarantine, or blocked by a tenant rule. A safe-sender setting or a user action can produce a different result from a provider-wide reputation problem.

Step 3: Check the sender and domain

Authentication is missing, broken, or misaligned

Google’s sender guidelines require all senders to use SPF or DKIM, valid forward and reverse DNS, TLS, and correctly formatted messages when sending to Gmail. Bulk senders, roughly 5,000 or more messages to personal Gmail accounts in 24 hours, must use SPF and DKIM, publish DMARC, align the From domain with SPF or DKIM, and support one-click unsubscribe for marketing and subscribed messages.

Check the actual DNS records and the header from the failed message. Possible breakpoints include:

an ESP was added to the workflow but not to the sending domain’s authorization;
a DKIM selector was deleted or rotated without updating the provider;
the visible From domain no longer aligns with the DKIM or SPF domain;
a copied SPF record has a syntax or lookup problem; or
the sending IP’s forward and reverse DNS do not match.

Fix the named record or sending service first. A copy edit cannot repair a failed authentication result.

Your list is producing bounces or complaints

Send only to people who asked for the messages and can recognize why they are receiving them. Old addresses, purchased lists, scraped contacts, unexpected frequency, and unclear opt-in can increase bounces or spam reports. Google recommends confirming addresses, periodically checking whether recipients still want mail, considering suppression of people who do not engage, and making unsubscribe easy.

Do not use a single bounce-rate benchmark as a universal pass/fail rule. Segment the evidence by source, age, campaign, and provider. Suppress hard bounces promptly, stop sending to addresses that repeatedly fail, and investigate a complaint spike before resuming full volume.

For Gmail, monitor the user-reported spam rate in Postmaster Tools. Google’s current FAQ says senders should keep it below 0.1% and prevent it from reaching 0.3%; rates at or above 0.3% have a greater negative impact on delivery for bulk senders. The metric is based on mail delivered to the inbox and reported by users, so a very low or suddenly zero rate can also mean Gmail is sending more of the mail directly to spam.

A new pattern or shared infrastructure changed the signal

A new domain, mailbox, sending IP, ESP, tracking domain, or campaign can change how a provider evaluates the stream. A shared IP also carries the activity of other senders; Google notes that a negative shared-IP reputation can affect delivery for every sender on that IP. Check which IPs and link domains appear in the headers, compare them with the last known-good send, and ask the ESP whether the pool or routing changed.

A public blocklist result is a clue, not a complete diagnosis. Check the domain, sending IP, and link domains separately, then follow the list owner’s current remediation process. Do not buy a new domain or request delisting until the behavior that caused the signal is understood.

Your marketing stream makes opting out difficult

For Gmail bulk senders, marketing and subscribed messages need one-click unsubscribe headers and a visible unsubscribe link. Google documents the List-Unsubscribe-Post: List-Unsubscribe=One-Click and List-Unsubscribe headers and recommends honoring requests within 48 hours. One-click unsubscribe is not required for transactional messages; classify each stream accurately.

Check the headers on the exact message that was filtered and test the unsubscribe endpoint from outside your application. A visible footer link alone is not the same as the provider’s one-click header requirement.

Step 4: Check the message and the recipient context

Misleading identity, links, or formatting

Gmail’s formatting guidance says headers and content must be accurate and not misleading. Display names should identify the sender, not imitate a reply, urgency notice, or recipient. Links should be visible and understandable, and hidden HTML or CSS can create risk. Microsoft’s headers can also reveal bulk or policy classifications.

Review the complete message, not just the subject:

Does the From name match the organization or person the recipient expects?
Do the visible links lead to the domains and destinations the copy describes?
Is the message readable when images are unavailable?
Are there unnecessary redirects, unexpected attachments, or a mismatch between the call to action and the landing page?
Does the HTML follow normal standards and include a useful text alternative?

There is no universal list of “spam words” that explains every placement decision. Test a plain, accurate version only after checking authentication, list quality, provider signals, and links. A word change is not a fix for a DNS or reputation problem.

Recipient or organization rules are involved

A message can be accepted by the provider and still be moved by a recipient’s settings, a company gateway, a quarantine policy, or an administrator rule. Gmail shows a warning or reason in some spam-folder cases; Microsoft recipients may need an administrator to inspect quarantine and tenant policy. Ask one affected recipient to check the message details and compare with a different recipient at the same provider.

Provider-specific checks

Gmail

Postmaster Tools covers messages sent to personal Gmail accounts such as @gmail.com and @googlemail.com; Google says its sender requirements and enforcement apply to those accounts, not Google Workspace inbound mail. Low-volume days may have missing data. Use the dashboard for user-reported spam rate, authentication, delivery errors, and compliance signals, but pair it with seed placement and direct recipient evidence for business domains. For setup and dashboard details, see Folderly’s Google Postmaster Tools guide.

When a dashboard goes quiet after a period of high spam, do not assume recovery. Google explicitly documents that automatic spam placement can reduce the number of inbox messages available for users to report.

Outlook.com and Microsoft 365

Use X-Forefront-Antispam-Report for message-level clues. CAT, SFV, and DIR help distinguish spam filtering, bulk classification, blocked or allowed sender rules, and other policy paths. If you control the sending IP, Microsoft’s current Smart Network Data Services portal provides IP data and access to the Junk Email Reporting Program. SNDS is IP-based; it is not a domain-wide report for every ESP customer. For related guidance, see Folderly’s Microsoft SNDS guide and Outlook spam filter guide.

Other providers

Provider dashboards are not interchangeable. Keep the provider, sending IP, authenticated domains, list segment, and message version with each placement observation. If a provider has no sender dashboard for your traffic, use full headers, bounce text, complaint data, and controlled seed tests rather than copying a Gmail conclusion onto another mailbox system.

A practical fix order

Use this order so each change leaves a clean signal:

Save one filtered message, one known-good message, their headers, the exact recipient providers, and the send timestamps.
Fix SPF, DKIM, DMARC alignment, forward/reverse DNS, TLS, and any provider error named in a bounce.
Pause the affected segment if complaints or hard bounces are rising. Confirm opt-in, suppress invalid addresses, and remove recipients who no longer want the stream.
Add and test the required unsubscribe headers for promotional mail; verify that requests are honored.
Compare sending IPs, ESP routes, tracking domains, volume pattern, and blocklist signals with the last known-good send.
Simplify misleading identity, links, attachments, and markup only after the sender and list checks are complete.
Send a controlled placement test and monitor each provider separately. Treat the result as evidence for that message, recipient set, and time, not as a permanent inbox guarantee.

For a first technical check, use Folderly’s live domain reputation checker and DMARC checker. They are diagnostic checks; no checker can override a recipient’s policy or guarantee inbox placement.

For a pre-send prevention checklist, see Folderly’s prevent emails from going to spam guide.

FAQ

Why are my emails suddenly going to spam?

Start with what changed: a DNS record, DKIM key, sending service, IP or tracking domain, list import, audience, volume pattern, or provider policy. Compare a new spam-foldered message with a recent inboxed message and read both headers. If the change affects one provider, use that provider’s evidence before changing the whole program.

Why is Gmail sending my emails to spam?

Check Authentication-Results, the Gmail spam-folder explanation, and Postmaster Tools. Confirm SPF or DKIM for all senders and the additional DMARC, alignment, unsubscribe, DNS, TLS, and spam-rate requirements that apply to bulk sending. Remember that Postmaster Tools reports personal Gmail traffic, not most Google Workspace recipients.

Why do my emails go to Junk in Outlook but not Gmail?

Outlook and Gmail use different signals and recipient policies. Inspect X-Forefront-Antispam-Report, especially CAT, SFV, and DIR; check SNDS if you control the sending IP; and ask the affected recipient whether a tenant rule or quarantine policy acted. Do not infer an Outlook diagnosis from a Gmail spam-rate graph.

How do I stop emails from going to spam?

Fix the highest-signal issue first: authentication and alignment, then list quality and complaints, unsubscribe handling, sending infrastructure, provider-specific policy, and finally message content and links. Verify each change with headers and a controlled placement test. There is no copy-only shortcut for a broken DNS record or poor sender reputation.

Do spam words send emails to spam?

Not by themselves in every provider or message. Individual wording can contribute to a message that already has other risk signals, but there is no universal word list that replaces header, list, reputation, and provider checks. Keep identity and links accurate, then test copy changes against the same sender and audience.

How long does it take to get out of spam?

There is no universal recovery timer. Authentication changes depend on the DNS and provider path; reputation and complaint signals depend on subsequent sending behavior. Google says compliance data can take up to seven days to reflect recommended fixes, and a clean dashboard day does not prove that inbox placement has recovered. Monitor the same providers and segments over time rather than promising an instant reset.

Sources

Sources checked: September 28, 2026. Human SME review: pending.

Why Are My Emails Going to Spam? A Diagnostic Guide