Mail-Tester: How to Read Your Email Spam Score

By Kam Low · Updated 2026-08-19

Mail-Tester

Mail-Tester is a free tool that gives an email a spam score out of ten by checking its authentication, content, and sending reputation after you send a message to a unique test address.

One instrument inside a wider preflight
Content score spam score "free" contributes +1 flagged, not critical
The wider preflight · seven checks
domain sender audience content score · this one authentication placement reputation
the score answers one narrow question: is the content itself a problem
A token adds a point, a clean total clears one suspect. A clean score means content is not your problem.

How it works

You send your email to the unique address Mail-Tester generates, and it returns a score out of ten with a breakdown: SPF, DKIM, and DMARC status, spam-trigger content, blacklist status, and formatting issues. It's fast, free, and the common first check.

Reading the score

Aim for 10 out of 10, and treat anything below 8 as worth fixing before a real send. The breakdown tells you what to fix, and authentication problems are usually the biggest single lever. For deeper, provider-specific placement, testing covers tools like Litmus and GlockApps.

Treat the number as one reading inside email deliverability as a whole. A broader spam test email covers the same content ground with more placement detail, and the blacklist line in the report is worth confirming against a direct Spamhaus listing check, since that is the list most receivers actually consult.

George, our CEO, has a blunt version of why the content checks matter: if you just send images, Gmail, Yahoo and Outlook cannot read them, so they file the message under promotions or spam. Alt text is what tells the inbox what the email is about.

The authentication bar underneath every score is concrete. Google's sender guidelines for anyone sending 5,000 or more messages a day say the DMARC enforcement policy can be set to none, provided the From domain is aligned with either the SPF domain or the DKIM domain.

And when the score is clean but placement is still bad, George's triage runs in order: authentication, then reputation, then content, then the AI-mediated inbox, Gemini summaries included.

Mail-Tester also assumes the message arrived. If it never reaches the test address at all, the problem is below the score, and a manual SMTP session over telnet shows exactly which command the server rejected.

Running a message through Mail-Tester by hand, reading the breakdown, and rewriting content one flag at a time is still manual work bolted onto a manual send. The real fix is having an AI agent send the email so it rarely trips those flags in the first place: describe the email in Claude, ChatGPT or Cursor, and Nitrosend's MCP server drafts it inside Brand Kit's own colors, fonts and tone, which is most of what a content score is actually checking for.

How Nitrosend compares

Mail-Tester scores one message against one test address. Nitrosend measures placement against a first-party pool of about twenty real seed mailboxes across Gmail, Workspace, Outlook and O365, Yahoo and Apple, and runs automated quality checks on every send, so testing is continuous and built into the send path rather than a manual step.

That difference matters most for transactional email, where receipts and password resets go out one at a time with no campaign moment to pause and paste into a scoring form. Sending through an email API means the scoring has to run inline on every message or it does not run at all.

Every campaign runs a preflight of seven distinct validation checks: domain verification, sender configuration, audience validity, content compliance and spam scoring among them, with dry_run validation for test sends. A recent campaign passed all seven with a spam score of zero.

That mirrors what the Email Marketing Bible tells any sender to run before sending: resolve wrapped CTA URLs, spam score, image-to-text weight, dark-mode lint, and a test to a seed address. Some tools build versions of this in. If yours does not, do it yourself.

The scoring stays legible too. One generated email scored 1 out of 10, low, with the word "free" contributing a single point that the report itself marked as not critical.

DNS checks cover SPF, DKIM, DMARC and return-path alignment for sending domains, surfaced as author_dns_health for visible-from domains.

Missing records are common in the wild. Nick on our team audited one domain-verification issue and found 13 accounts that were not sending from the address they wanted, purely from missing records.

George wants to push placement testing further still: a large network of agent-controlled inboxes across all the inbox providers, to improve monitoring, work out how to hit the primary tab, and feed better analytics back into reports.

FAQ

What is Mail-Tester?

A free tool that scores an email out of ten by checking authentication, content, and reputation after you send a message to a unique test address.

What is a good Mail-Tester score?

Aim for 10 out of 10. Below 8 is worth fixing before sending to your list, and the breakdown shows exactly what to address.

Is Mail-Tester enough to test deliverability?

It's a strong quick check for one message. For provider-specific inbox placement across a program, add seed-list testing.

Related reading

Transactional email, defined

The transactional email guide

Webhooks explained

Email validation in the API