Short answer: spam trigger words matter far less than the advice suggests, and swapping “free” for “no cost” will not move an email out of the spam folder. No major mailbox provider publishes a list of forbidden words, and none of them decide placement by scanning your copy against one. They weigh the sender: whether your mail is authenticated, how many recipients mark it as spam, whether the addresses on your list are real, and whether this send looks like the ones before it.
Content is not weightless, and I will quote Google on exactly what still counts. But the published rules turn on honesty and message structure, and they barely touch vocabulary.
Where the spam trigger word lists came from
They were not invented out of nothing. They are just old.
Filters in the late 1990s and 2000s really did score content. Broadly, a rule engine added points for suspicious phrases, odd formatting and header problems, and anything over the threshold got flagged. That is why “act now”, “risk free” and a row of exclamation marks became famous.
Then spammers, who had the strongest incentive of anyone, learned to rewrite around the rules faster than anyone could maintain them, which left the rules mostly catching honest marketers who had never read them. Providers meanwhile got something better: billions of messages a day, and a record of what recipients did with each one. Delivered, opened, replied to, deleted, reported. A thesaurus cannot fake any of that.
The lists outlived the filters that produced them, and they are still republished today with the same examples from twenty years ago.
Why spam trigger words stopped deciding placement
Modern filtering is sender-first. Before a message is judged on its own merits, the receiving server already knows a lot about where it came from: your domain’s reputation (a running score mailbox providers keep for your sending domain and sending server), whether your authentication records line up, and how recipients have treated your mail recently.
That order explains the thing everyone notices and nobody can account for. A sender with a clean history puts “free” in a subject line and lands in the inbox. A sender with a damaged domain avoids every word on every list and still lands in spam.
You can see part of this for yourself: Google and Yahoo expose pieces of the picture through free tools, and our guide on how to check your sender reputation shows where to look. Both also enforce requirements for bulk senders, covering authentication, complaint rates and unsubscribe handling. None of those requirements mentions which words you may use.
What content still counts for
Google publishes sender guidelines for Gmail, and they do include content rules. They are much narrower than people expect. Quoting the page directly:
- “Email message subject, headers, display names, and other message elements should accurately represent the sender identity and message content, and shouldn’t be misleading.”
- “don’t send messages with subject lines starting with Re: or Fwd: unless the messages are actual replies or forwards”
- “Don’t use HTML and CSS to hide content in your messages. Hiding content might cause messages to be marked as spam.”
- “Web links in the message body should be visible and easy to understand. Recipients should know what to expect when they click a link.”
- “Don’t mix different types of content in the same message. For example, don’t include promotions in sales receipt messages.”
- “If your messages are in HTML, format them according to HTML standards.”
Every item there is about whether a message misrepresents itself or is put together badly. No vocabulary guidance anywhere.
Two other patterns are widely reported by deliverability practitioners to travel with worse placement. Treat them as folklore, since no provider documents them:
- An email that is one large image with a link and barely any text. Filters cannot read the image, and the shape matches a known spam pattern.
- Heavy use of link shorteners, because the destination stays hidden until the click. Google’s page does not mention shorteners at all, so this one is inference.
Both are structural. You fix them by changing how the message is built.
Common claims about spam trigger words, and what actually happens
| The claim | What actually happens |
|---|---|
| “Never put free in a subject line” | From a sender with a healthy reputation and a real list, “free” reads as ordinary marketing language and goes through. Retailers use it every week. |
| “Urgent will get you filtered” | The word is fine. “Urgent” attached to a claim that is not true is misrepresentation, and that is what Google’s guidelines warn about. |
| “Exclamation marks and ALL CAPS trigger filters” | Neither is documented as a rule by the major providers. Both can push a reader toward the report spam button, so any damage is indirect. |
| “Never write click here” | Google asks for links that are visible and easy to understand, so a vague label is a small content issue. It is a wording problem, not a filter rule. |
| “Prices and dollar signs are fatal” | Sales emails with prices in the subject reach inboxes all day. |
What to check instead of your word choice
If you have spent a week rewriting subject lines, here is where that time is better spent. Roughly in the order the fault usually turns out to be.
- Authentication, and alignment. SPF, DKIM and DMARC are the DNS records that let a receiving server confirm a message really came from your domain. Having them is not enough: the domain in your visible From address has to match the domain that passed. If you send in volume, the Gmail and Yahoo bulk sender rules set the baseline.
- Your spam complaint rate. The share of recipients who hit “report spam”. Google publishes a ceiling for it in Postmaster Tools, and it is the number most likely to explain a sudden drop. Our post on what a healthy spam complaint rate looks like covers where to find yours.
- How old the list is, and where it came from. Addresses collected at a trade show two years ago carry a different risk from addresses that opted in last month. If nobody can say when a segment was collected, you have probably found your answer.
- Whether the addresses still exist. Mailboxes close constantly as people change jobs and abandon accounts. Every dead address becomes a bounce, and a run of bounces tells the provider you do not know your own list.
- Whether unsubscribing actually works. A broken or buried unsubscribe link converts people who wanted to leave into people who report you. Bulk marketing mail is expected to offer one-click unsubscribe.
- Whether your sending stays steady. A domain that sends a few hundred messages a day and then pushes out a hundred thousand with no warning looks like an account somebody has taken over.
If those check out and mail is still being filtered, our walkthrough of why emails go to spam even when the content is fine takes the diagnosis step by step.
An honest opinion
The word lists survive because they are the only deliverability advice that costs nothing and needs nobody’s permission. You can rewrite a subject line in ten seconds. You cannot fix a DNS record without access, or clean a list without admitting it needs cleaning. Rewriting feels like progress, and mostly it is displacement activity while the real damage carries on.
There is one honest use for the lists. If a subject line lights up half of one, that is often a decent signal the message really is overheated, and your readers will feel it long before any filter does. Take a high score on a spam-word checker as a note about your writing, and ignore its verdict on your deliverability.
The part of your list that does more damage than any word
Of everything on that checklist, the item I would fix first is the one most people skip: whether the addresses are real.
Dead addresses bounce, and bounce rates feed reputation directly. Spam traps are worse: a trap exists only to catch senders, so mail arriving there says something about how you built your list that no careful copywriting can talk you out of. Changing your vocabulary touches neither problem.
An email verifier checks each address against the receiving mail server without sending a message, then reports whether the mailbox exists. Running Reoon Email Verifier over a list before a campaign removes the addresses that would have bounced, flags known traps, and marks disposable addresses, meaning the temporary mailboxes people create for one sign-up and never open again. It will not repair a reputation that is already damaged, but it does stop the damage that repeats with every send.
If you want to see what your list looks like before the next campaign, a free Reoon account comes with free verification credits every day, no card required. Verify a sample and you will know within minutes whether your copy was ever the problem.
FAQ
Is there an official list of spam trigger words?
No. No major mailbox provider publishes one, and the lists circulating online are not drawn from Gmail, Outlook or Yahoo documentation. They descend from the rule-based content filters of the 2000s and from checker tools that invented their own scoring. Google’s sender guidelines cover misleading subject lines and hidden content, but never name a forbidden word.
Does using “free” in a subject line send an email to spam?
On its own, no. Retail and newsletter mail with “free” in the subject reaches inboxes every day, because those senders are authenticated and their complaint numbers stay low. If your mail is being filtered and the subject happens to say “free”, the word is a symptom rather than a cause. Check reputation and list quality first.
Do exclamation marks and capital letters hurt deliverability?
Not as a documented filter rule at the major providers. The indirect effect is real, though. A subject line shouting in capitals with three exclamation marks reads as pushy, and readers who feel pushed are the ones most likely to hit “report spam” instead of simply deleting. Complaints do count against your sending reputation, so the punctuation hurts you through your readers.
What should I change first if my emails keep going to spam?
Authentication. Send a test message to a Gmail account you control, open “Show original”, and confirm that SPF, DKIM and DMARC all pass with your own domain beside each result. If they pass, look at your spam complaint rate next, then at how old your list is. Subject line wording belongs near the bottom of that queue.







