An email spam checker tries to answer one question before you send: will this message land in the inbox or in the spam folder? No tool can promise the answer, because the final decision belongs to the recipient's mailbox provider, and each provider uses its own signals. What you can do is check the things that you control, test with real mailboxes, and watch the results over time. This guide explains what spam filters look at, what a spam checker can and cannot tell you, and a practical way to test an email before it goes out.
How spam filters decide
Filters at large mailbox providers combine many signals. Public documentation from the main providers highlights three groups. The first is authentication: whether SPF, DKIM and DMARC pass, and whether the domains align. The second is reputation: the history of your domain and sending IP address, including complaints, bounces and blocklist entries. The third is content and recipient behaviour: how the message looks, whether people open it, reply to it, move it out of spam, or mark it as spam. Providers do not publish exact formulas, and a score from one tool is never the same as the decision of a real provider.
What a spam checker can tell you
- Authentication results. Whether SPF, DKIM and DMARC are set up and whether they pass for a test message.
- Technical errors. Missing headers, broken HTML, a missing plain-text part, a missing unsubscribe header, mismatched domains in links.
- Blocklist status. Whether your sending IP or domain appears on common lists.
- Content flags. Phrases, formatting or link patterns that tend to trigger filters, such as an image-only message or a link shortener.
What a spam checker cannot tell you
- Where the message will land at Gmail, Outlook or Yahoo. Only a real send to real mailboxes shows that, and the result can differ between recipients.
- How your audience will react. Complaints and engagement are decisive, and no tool sees them in advance.
- The long-term effect of your sending history. A clean test message does not repair a poor reputation.
Treat any "spam score" as a helpful hint. A score of "good" does not mean the message will reach the inbox, and a poor score points to things worth fixing. Be careful with tools that promise a guaranteed inbox.
A practical way to test an email before sending
- Check your domain. Run the SPF, DKIM and DMARC checker and fix any failure. Our guide to authenticating your sending domain shows the steps.
- Check blocklists. Use the blacklist check for your sending IP and domain. If you appear on a list, read the delisting guide.
- Send test messages to real mailboxes. Create test addresses at the major providers and at your business email system, send the actual message, and look at where it lands. Do this for each template, not once. See inbox placement testing for how seed lists work and where they mislead.
- Read the headers of a delivered test. The authentication results show what the receiving server saw. See how to read email headers.
- Review the content. Use a clear subject, a real From name, a plain-text version, working links on your own domain and a visible unsubscribe link. See email content and spam filters.
- Start small if the list is new. Send to your most engaged contacts first and increase gradually. See the email warm-up guide.
Common content problems that trigger filters
| Problem | Why it hurts | Better practice |
|---|---|---|
| One large image and almost no text | Filters cannot read it, and it looks like evasion | Balance text and images, and write useful alt text |
| Link shorteners or mismatched link domains | Often used to hide destinations | Use full links on a domain you control |
| Misleading subject or From name | Drives complaints | Say honestly what the message is and who it is from |
| No plain-text part | Looks automated and careless | Send multipart/alternative with both parts |
| No unsubscribe link or header | Pushes people to the spam button | Offer a visible link and a one-click unsubscribe header |
| Sudden large volume from a new sender | Looks like abuse | Increase volume gradually |
Monitoring after you send
Testing before you send is only half of the work. After a send, look at bounces, complaints, blocks and engagement for each major provider. A spike in one place points to the cause. For Gmail, Google Postmaster Tools shows domain reputation and spam rate. Our guides on complaint rates and deliverability metrics explain what to look at. If a message you sent has already gone to spam, follow the spam troubleshooting checklist.
Beware of "spam word" lists. Using a certain word does not, by itself, send an email to spam. Modern filters weigh authentication, reputation and recipient behaviour far more heavily than any single phrase. Write for your reader, avoid deception and the rest tends to follow.
If you need a deeper review
When tests keep failing and you cannot see why, a structured review of your authentication, infrastructure, list and content finds problems that a quick check misses. Our deliverability audit is designed for that purpose.
Using seed tests without fooling yourself
A seed test sends your message to a small set of mailboxes and reports where each one landed. It is useful, and it has limits you should understand. Seed mailboxes have no history with your sender, so they do not behave like your real subscribers, who open, reply and delete in ways that shape placement. A result of "inbox" at a seed address does not guarantee the same for a customer who never opens your mail. Use seed tests to spot sudden changes, such as a template that suddenly lands in spam, and not as proof of inbox delivery. Compare results for each provider over time, and treat one bad result as a prompt to investigate, not as a verdict.
Build a pre-send checklist
Turn the steps above into a short checklist that your team follows for each new template and each large campaign: authentication passes, the sending IP and domain are not listed, a test reached a real inbox at each major provider, the plain-text part exists, links go to your own domain, the unsubscribe link and header work, and the audience is made up of recent, consented contacts. A checklist is dull, and it prevents most avoidable problems. Keep a record of each result so that you can see when something changed.
What to do when the test says spam
Do not guess. Change one thing at a time and test again. Start with authentication, since a failure there outweighs everything else. Then look at the sending history: has the volume jumped, or have complaints risen? Then look at the content and the audience. If a single provider is the problem, look at that provider's own tools and guidelines. If all providers put the message in spam, the cause is usually authentication, a blocklist or a list problem. Slow down and send only to your most engaged contacts while you fix the cause.
Check your domain before you send Run the free SPF, DKIM and DMARC checker to see whether your authentication is set up correctly.
References and further reading
Frequently asked questions
What is an email spam checker?
It is a tool that tests a message or a domain for the things that commonly cause spam filtering, such as failed authentication, blocklist entries, broken HTML and risky content patterns. It gives hints, not a guarantee.
Can a spam checker guarantee inbox delivery?
No. The decision belongs to the recipient's mailbox provider, and it uses signals no tool can see, such as your reputation and how recipients react. Real test sends show actual placement.
How do I test if my email will go to spam?
Check SPF, DKIM and DMARC, check blocklists, send the real message to test mailboxes at major providers, read the headers of a delivered test, review the content and monitor results after sending.
Do spam trigger words send email to spam?
Not on their own. Modern filters weigh authentication, reputation and recipient behaviour much more than single words. Misleading content and poor list quality matter far more.
What should I do if my email already went to spam?
Check authentication and blocklists, review recent complaints and bounces, look at list quality and content, and reduce volume to your most engaged contacts until placement recovers.