Email Spam Checker

Problem: Why Did My Email Land in Junk?

Mail servers do not decide what is spam by reading your message and forming an opinion. They score it. Every commercial filter runs a rule engine over the raw message and adds up points, and once the total crosses a threshold the message goes to the junk folder no matter how good the offer is. The classic example is SpamAssassin, which has shipped with a pass mark of 5.0 out of 10 for twenty years. Gmail, Outlook and Yahoo all run heavier, machine-learned versions of the same idea now, but the underlying mechanic has not changed: you accumulate risk, and then a line is drawn. This tool runs that same accounting on your copy before you send it, so the score is something you can argue with instead of something you discover afterwards in a seed test.

Why the Subject Line Is Scored Separately

Two things make a subject line more dangerous than the body. It is the only part most recipients read before deciding whether to delete, so filters weigh it harder, and it is deliberately deceptive in a specific, legally recognised way. Writing a reply prefix you never earned is listed as a deceptive subject line under CAN-SPAM, and it is the single most reliable marker of bulk spam. On top of that, a subject that opens with a dollar sign or a percentage gets cut short on a phone, so the words that sell the email are the words a reader never sees. This tool therefore scores the subject on its own, applies the trigger words at double weight there, and separately checks the hidden preview text that fills the gap next to it in the inbox list.

What It Does Not Know

There is a limit to what a page can know, and it is worth being precise about it. This tool reads the copy you paste. It cannot see the sending IP, so it cannot check blocklists, and it cannot judge how many complaint reports your domain has collected. It also cannot open your DNS, so SPF, DKIM and DMARC are outside its reach, and a real inbox placement test means actually sending a message to a seed list and waiting for Gmail to sort it. All of that is server-side work. What is genuinely reachable from a browser is the copy, and the copy is where most of the damage is done, because it is the part you actually control and the part you can change today.

Every Finding Comes With a Fix

A finding without a rewrite is just criticism. Every phrase this tool flags ships with the wording the industry recommends instead, and the fix button rewrites the whole message in one pass so you can paste the corrected version straight back into your campaign. It is worth expecting a couple of hits even on copy you are happy with. Trade language is what marketers use, and filters are deliberately blunt about money and urgency because the overwhelming majority of email using those words is fraud. A finding is a question, not a verdict, and the score is a temperature reading rather than a judgement.

Specificity Is the Whole Trick

Filters reward specificity, and that is the whole trick. A newsletter that says the seminar runs on the fourteenth at nine in the morning Eastern gets read, because there is nothing in that sentence for a rule engine to object to. The same newsletter rewritten as do not miss this amazing limited time opportunity gets scored on sight. Nothing about the offer changed, but one version sounds like a person and the other sounds like a list. Keeping numbers, dates and names in place of adjectives is the cheapest lift available to anyone who sends email, and it costs nothing but a rewrite.

How the Score Is Built

• 552 trigger phrases in 16 categories, each carrying a weight taken from the risk rating the phrase is published with: high, medium or low.
• Double weight in the subject line, because that is where filters look first.
• Per-category caps, so one repeated word cannot push a message over the line on its own.
• Structural rules for things no word list can catch: hidden text, fake reply prefixes, ALL-CAPS runs, bare URLs, anchor text pointing somewhere other than where it claims, and images with no alt text.
• Compliance checks for the CAN-SPAM elements that get businesses fined, which is the part most free checkers leave out entirely.

Frequently Asked Questions

What score should I be aiming for?

Below 2.0 is clean. Between 2.0 and 4.9 is borderline, which is where a lot of legitimate marketing email sits and where you should look at the top findings rather than the number. 5.0 and above is the classic SpamAssassin pass mark and means a filter running that rule set would treat the message as spam. Above 8.0 the message is almost certainly being bulked before anyone reads it.

Is a high score a guarantee it will be filtered?

No, and the reverse is not a guarantee either. This is a transparent model of how rule-based filters score, not the filter itself. Gmail and Outlook both add machine-learned models on top that weigh engagement history, sender reputation and recipient behaviour, none of which a page can see. Treat the score as a floor: get it down, and you remove the part you can actually control.

Why does the word "free" cost so much when my newsletter mentions it once?

Because a rule engine cannot tell your one innocent mention from the several hundred marketing emails that use "free" as the entire pitch, so it charges the same either way. The fix is rarely to delete the word. It is to put something specific next to it, so the sentence stops reading like an advert, for example "your first 30 days free, no card needed".

Does this send my email anywhere?

No. There is no network request in this tool. The scoring runs in JavaScript in your browser and nothing is stored, logged or sent. That is the main practical difference from a seed tester such as Mail-tester or GlockApps, where you hand a copy of your email to a third party so they can watch where it lands.

My email already passed in a seed test. Why is this higher?

Two reasons. A seed test usually scores the delivered message after the receiving server has added its own headers, which is a different input from your raw copy. And seed tests report the filter's final verdict, not its arithmetic, so a passing 4.8 and a failing 5.1 look identical from outside. If you want the reasoning rather than the verdict, this is the tool.

Can I paste HTML instead of plain text?

Yes, and you should if you have it. Paste the raw HTML of the email and the tool reads the visible text, the links with their anchor text, the images and their alt attributes, and flags anything hidden with display:none, zero font size or white text on a white background. Hidden text is treated as a deceptive practice rather than a style choice, because that is what it almost always is.

Does it check CAN-SPAM compliance?

It checks the elements that live in the copy: a valid sender identification, a non-deceptive subject line, a working unsubscribe path and a physical postal address. It cannot check that your unsubscribe link actually processes the request, which is the part that matters most in practice. Under CAN-SPAM each separate email that fails is subject to a fine, and enforcement since 2024 has been carried out largely by universities rather than the FTC.

How large is the trigger list, and where does it come from?

552 phrases across 16 categories, each with the risk rating it is published under and a recommended alternative. The list is embedded in this page as a static data block, so the tool works offline and the scoring is identical for every user and every run. Nothing about it changes between visits, which is what makes the number comparable over time.