Skip to main content
Free tool

Cold email spam word checker

Paste a subject line and body. See which phrases and patterns look like bulk promotional mail, and how to rewrite them.

Runs in your browser. Nothing is uploaded.
Your score, flagged phrases and fixes will appear here.

How to use the checker

  1. Paste the subject line you plan to use. Subjects are checked more strictly than the body, because they are the first thing both filters and people read.
  2. Paste the body exactly as it will be sent, after personalization if you can. Plain text and HTML both work.
  3. Select Check email. Work through the "What to fix" list from the top; the highest-severity items cost the most points.
  4. Edit and check again until the remaining flags are ones you have consciously chosen to keep.

What the results mean

The score starts at 100 and each issue subtracts the points shown next to it, so you can see exactly why a message scored the way it did. Grades run from A (90 and above) to F (below 60). The checks fall into five groups:

  • Trigger phrases. Wording that shows up disproportionately in unsolicited promotional mail: money claims ("make money", "guaranteed income"), pressure ("act now", "last chance"), and hard calls to action ("click here", "buy now"). Matching is by whole word, so "freedom" does not count as "free". Each flagged phrase comes with a suggested rewrite.
  • Subject patterns. All-caps subjects, exclamation marks, dollar signs and percentages, piles of emoji, and fake "RE:" or "FW:" prefixes on a first email. A fake reply prefix is also deceptive, which CAN-SPAM prohibits.
  • Links. The number of links, URL shorteners (which hide the destination and are heavily abused), domain endings commonly associated with abuse, raw IP addresses, and unencrypted http:// links.
  • Formatting. Images with little supporting text, heavy HTML, shouting in capitals, stacked exclamation marks, and broken merge tags such as {{firstName} or a literal "[First Name]".
  • Compliance. Whether there is any way for the recipient to opt out. A simple line such as "reply 'no thanks' and I won't follow up" counts; commercial email in the US also needs a postal address.

Why content is only part of the picture

Modern spam filtering relies far less on keyword lists than it did a decade ago. Gmail, Outlook and Yahoo look mainly at who is sending: whether the domain is authenticated, how its past mail was received, and whether recipients mark it as spam. Content matters most at the margins, for a new domain without much history, or when a message genuinely reads like bulk promotion.

So treat this score as a writing check, not a delivery forecast. If your messages land in spam even with a clean score, check the sending domain's SPF, DKIM and DMARC records with the email deliverability checker, and check your list with the email address checker before sending, since bounces hurt sender reputation too.

Writing cold email that reads like a person wrote it

  • Keep the subject short, specific and in sentence case. "Question about your Denver office" beats "Exclusive offer inside!".
  • Open with why you are writing to this person, not with what you sell.
  • Make one ask, ideally a question that can be answered with a reply. Use one link at most, on your own domain.
  • Skip images, attachments and heavy templates in a first email.
  • Give an easy way out. People who can say no politely are less likely to press "Report spam".

For more on the sending side, see common email deliverability mistakes.

FAQ

Questions

No. This checker only looks at the words and formatting of one message. Mailbox providers weigh sender reputation, authentication (SPF, DKIM, DMARC), complaint rates and how recipients engage far more heavily than any single phrase. A clean message from a new or poorly authenticated domain can still land in spam.

One instance will not sink a message on its own. The risk comes from stacking promotional phrases, and from vague offers that read like bulk marketing. The checker flags each phrase so you can decide; in cold email, a specific description of what you are offering usually reads better anyway.

No. The analysis runs entirely in your browser. Nothing you paste is uploaded, stored or logged by AngelSend.

Yes. HTML is converted to text before phrases are matched, and the checker also looks at the text-to-markup ratio and image count. For cold outreach, plain text is generally the safer format.

In the US, CAN-SPAM requires commercial email to include a way to opt out and a valid postal address. Beyond the law, an easy way to say no reduces spam complaints, which mailbox providers do track.

More free tools

These tools are free and run in your browser. They are made by AngelSend, an agentic CRM that sends outreach from your own mailbox and runs these kinds of checks before each send.