What words trigger spam filters?

Mostly phrases rather than single words, and only in rule-based filters like SpamAssassin, which scores patterns such as "100% guaranteed", "money back guarantee" and "Dear Friend". Gmail and Outlook use learned models and sender reputation and publish no word list, so no single word is banned there.

Is there an up-to-date spam trigger words list for 2026?

Not one from a mailbox provider. The long lists online are compiled from marketing advice, not from filter rules, and none of them is published by Gmail, Outlook or Yahoo. The closest thing to an official list is SpamAssassin's published rules, which this checker uses.

Is "free" a spam word?

Not on its own in any published filter rule. SpamAssassin scores specific phrases built around it, such as a "free instant quote", and words disguised to dodge filters, like f-r-e-e. The word appears on cold email trigger lists because it makes an email read like an ad, not because a filter penalises it.

Do spam trigger words matter more in the subject line?

SpamAssassin has separate rules for subjects, such as all capitals or a subject that starts with a dollar amount, so a subject is checked on its own. Run it through the subject line tester, which uses those rules and previews where the line gets cut off.

How does this checker work?

It runs in your browser. Your text is matched against phrase rules ported from SpamAssassin 3.3, checks drawn from Google's sender guidelines (link shorteners, fake Re: prefixes) and a labelled list of common trigger words. Each finding says which of the three it is.

Why doesn't it give a spam score?

Because a score from the text alone would be invented. Real filters combine content with sender reputation, authentication, recipient feedback and learned models that no outside tool can see. We show the published SpamAssassin points per rule so you can judge them, without adding them up into a verdict.

My copy is clean but emails still go to spam. Why?

Usually reputation or authentication, not words. Check SPF, DKIM and DMARC, your spam complaint rate in Google Postmaster Tools, and your bounce rate. Mailing addresses that don't exist, or spam traps, lowers reputation for every message that follows.

Email spam words checker, with the rule behind every flag

Single words rarely send an email to spam. Gmail and Outlook use learned models and sender reputation and publish no word list. Rule-based filters such as SpamAssassin do score phrases like "100% guaranteed" or "Dear Friend". This spam words checker marks those, flags what breaks Google's sender guidelines, and labels the rest as common advice.

Your email copy (subject and body, plain text)

Checked in your browser. Nothing is sent anywhere.

SimpleAPI

Sample result

Your copy has 6 findings, including 3 SpamAssassin rules and a link shortener.

Matches: 6 findings
Google guideline: Link shortener
Highest score: EXCUSE_REMOVE, 3.299

What this means for you

Swap the flagged phrases and use full links on your own domain before you send.

Next step:Check the list you're sending to

SpamAssassin 3.3 rules, published scores
Google sender guideline checks
Trigger words labelled as practice only
A suggested swap for each match

Spam words
SpamAssassin
Copy

By SimpleVerifier Team · Last updated October 17, 2026 · Runs in your browser; nothing you paste is sent to us

How do I check an email for spam trigger words?

Paste the email you're about to send, subject line included, as plain text.
Press Check the copy. Matches are marked in your text.
Read the label on each finding. SpamAssassin rules show the project's published score; Google guideline issues are worth fixing outright; common-practice flags are judgement calls.
Use the suggested swaps where they make the email clearer. Then check the subject on its own with the subject line tester.

Example

You enter: A promotional email with “ACT NOW”, a money-back style guarantee, a bit.ly link and a “To be removed from our mailings” line.
You get: 6 findings: 3 SpamAssassin rules with their published scores (highest EXCUSE_REMOVE at 3.299), a Google guideline flag for the link shortener, and 2 common trigger words, each with a suggested swap.
Limits: No limit; runs in your browser.

Which spam trigger words actually matter?

The ones that match a published filter rule or break Google's guidelines; the rest are judgement calls. SpamAssassin rules are the only word-level rules with published weights. The open-source filter publishes each rule's pattern and default score, and marks a message as spam when the total reaches 5, a threshold its configuration docs call "quite aggressive". The phrases checked here are ported from its 3.3 rule files, with scores from the 3.3 rule list for a setup with learning and network tests on: MONEY_BACK ("money back guarantee") is 1.232 points, DEAR_FRIEND is 2.577. Any administrator can change or disable them, and newer releases differ, so treat them as a guide to the patterns filters look for, not a forecast.

Google guideline issues come from Google's Email sender guidelines, which ask for links that are "visible and easy to understand" and subjects that aren't misleading, such as a fake Re: or Fwd:. They aren't word rules, but they describe what Gmail's filters are built to catch, so they are the flags to fix first.

Common-practice flags are words like "free", "act now" and "click here" that appear on cold email trigger-word lists. No major filter publishes a penalty for any of them on their own. They're shown because they often make an email read like an ad, and the suggested swap is usually just clearer writing.

Does Gmail have a list of spam trigger words?

No. Google publishes no word list, and nothing in the text alone can predict where an email lands, which is why this checker shows no score out of 100. Google described its Gmail filtering in terms of machine-learned models, image-based messages, hidden content and newly created domains, not vocabulary. Microsoft Defender assigns a spam confidence level from its own filtering, which doesn't publish word weights either.

Even in SpamAssassin, the points that tip a message over are often not about wording. Its rule list gives 3.335 points for mail relayed through an IP on the Spamhaus PBL and 1.725 for a link to a blocklisted domain, both more than most phrase rules. Our post on the spam trigger words myth works through the numbers.

What actually keeps emails out of spam?

Sender reputation, built from how recipients treat your mail, and the quality of your list. Google asks senders to keep reported spam below 0.3%, and hard bounces from addresses that don't exist tell mailbox providers the list wasn't collected carefully. Spam traps, addresses that exist only to catch senders who mail without permission, are worse. No rewording fixes those.

So tidy the copy here, check authentication with the bulk sender checker, and make sure the addresses you're sending to exist before the email goes out. If mail is already landing in spam, why emails go to spam walks through the causes in order of impact.

Is a filter rule the same as a spam trigger word?

	SpamAssassin rule	Google guideline	Trigger word list
Published by	The SpamAssassin project	Google	Blogs and vendors
Has a weight	Yes, a score per rule	No, but describes what Gmail targets	No
Example	`DEAR_FRIEND`, 2.577 points	A fake Re: in the subject	"free", "act now"
Should you fix it?	Usually; the rewrite reads better too	Yes	Judgement call

Who uses this tool, and for what?

Cold emailers: Check each step of a sequence before it goes live. Opt-out lines like "to be removed from our mailings" match published rules; a plain "reply no and I won't follow up" doesn't.
Marketers: Paste a newsletter or promo as plain text to catch phrases such as "100% guaranteed" or "Act Now", then fix the subject separately with the subject line tester.
Agencies: Show a client why a phrase gets flagged, with the rule name and its published points, instead of arguing from a trigger-word list.
Sales ops: Review shared sales templates once, so every rep sends copy that has no published-rule matches and no fake Re: subjects.

What can’t this tool tell you?

It can't predict inbox placement. Reputation, authentication and recipient behaviour outweigh wording, and none of them is visible in your text.
Rules are ported from SpamAssassin 3.3. Newer releases and individual servers use different rules and scores.
It reads plain text. HTML problems such as hidden text, image-only emails or a mismatched link target aren't checked.
The patterns are English. Copy in other languages gets few or no matches, which doesn't mean it's clean.

Common questions

What words trigger spam filters?: Mostly phrases rather than single words, and only in rule-based filters like SpamAssassin, which scores patterns such as "100% guaranteed", "money back guarantee" and "Dear Friend". Gmail and Outlook use learned models and sender reputation and publish no word list, so no single word is banned there.
Is there an up-to-date spam trigger words list for 2026?: Not one from a mailbox provider. The long lists online are compiled from marketing advice, not from filter rules, and none of them is published by Gmail, Outlook or Yahoo. The closest thing to an official list is SpamAssassin's published rules, which this checker uses.
Is "free" a spam word?: Not on its own in any published filter rule. SpamAssassin scores specific phrases built around it, such as a "free instant quote", and words disguised to dodge filters, like f-r-e-e. The word appears on cold email trigger lists because it makes an email read like an ad, not because a filter penalises it.
Do spam trigger words matter more in the subject line?: SpamAssassin has separate rules for subjects, such as all capitals or a subject that starts with a dollar amount, so a subject is checked on its own. Run it through the subject line tester, which uses those rules and previews where the line gets cut off.
How does this checker work?: It runs in your browser. Your text is matched against phrase rules ported from SpamAssassin 3.3, checks drawn from Google's sender guidelines (link shorteners, fake Re: prefixes) and a labelled list of common trigger words. Each finding says which of the three it is.
Why doesn't it give a spam score?: Because a score from the text alone would be invented. Real filters combine content with sender reputation, authentication, recipient feedback and learned models that no outside tool can see. We show the published SpamAssassin points per rule so you can judge them, without adding them up into a verdict.
My copy is clean but emails still go to spam. Why?: Usually reputation or authentication, not words. Check SPF, DKIM and DMARC, your spam complaint rate in Google Postmaster Tools, and your bounce rate. Mailing addresses that don't exist, or spam traps, lowers reputation for every message that follows.

Sources

Authentication and capacity are half of deliverability.

The other half is who you send to. Every hard bounce counts against the domain and inboxes you are sending from, and nothing in DNS can tell you which addresses on your list are dead.

Bulk verification with real SMTP mailbox checks
Hard bounces removed before a send, not discovered after
Catch-all, disposable and role detection on every row

Get started — $29.99/mo See the published limits first

One plan, cancel anytime, no contract. Sign in with Google and your first 100 email checks are free, no card. The free tools stay free either way.