What does an email deliverability test check?
An email deliverability test sends one real message from your mailbox to an address that reports back on it. Free testers check your setup: SPF, DKIM, DMARC, blocklists and the message itself. Placement tests show which folder the email reached in test mailboxes. Neither shows how a provider treats mail to all your real prospects; provider dashboards and replies come closer. Each layer answers a different question, so run all three.
| Layer | Question it answers | Free way to run it | What it misses |
|---|---|---|---|
| Setup test | Is the email signed, authorised and sent from an IP that isn't blocklisted? | mail-tester.com, the MxToolbox ping test, reading the headers yourself | Which folder it lands in at Gmail or Microsoft |
| Placement test | Which folder did it reach at each provider? | A personal Gmail address and a Microsoft 365 work address that colleagues own | Test mailboxes aren't your prospects |
| Live monitoring | How does the provider rate all the mail you send? | Google Postmaster Tools, Yahoo's Complaint Feedback Loop | Needs volume; Postmaster Tools covers personal Gmail accounts only |
Before you send anything, confirm the records exist. Our free SPF checker reads your domain's SPF record and flags a second SPF record, too many lookups, +all at the end or no record at all.
How do you run a free deliverability test?
Send the exact email you plan to use, from the same mailbox and sending tool, with the same signature and links. A test from another mailbox tests another sender. These steps are for a desktop browser, from each tool's help page checked on 4 November 2026.
Send it to mail-tester
Open mail-tester.com, which generates a random address each time you visit. Send your email to it, then click Check your score. Its FAQ says it uses a default SpamAssassin install, with the score reversed so positive is good, and allows 3 free tests within 24 hours. Free results stay visible for 7 days to anyone with the test address, so don't test anything confidential.
Send it to MxToolbox
Send the same email to [email protected]. It replies with a link to your full Deliverability Report. MxToolbox's deliverability page says the report analyses your headers, the blacklist reputation of your outbound IP address and your SPF records.
Send it to two inboxes you don't own
Send to one personal Gmail address and one Microsoft 365 work address belonging to a colleague or friend, and ask them to tell you which folder it reached before they open it. Our cold email spam fix list reports commenters' view that your own inboxes make placement look better than it is.
Copy the full headers
In Gmail, open the email, click More next to Reply, then Show original, then Copy to clipboard (Gmail Help). In Outlook on the web or Outlook.com, select More actions, then View, then View message details. In classic Outlook for Windows, open the message, click File, then Properties, and copy the Internet headers box (Microsoft Support).
Paste them into an analyser
Gmail's help page points to Google Admin Toolbox Messageheader: paste the header and click Analyze the header above. For headers from a Microsoft 365 mailbox, Microsoft's documentation suggests the Message Header Analyzer, which lays the fields out in a readable form.
How do you read the test results?
Look for three passes and one spam verdict. The values below come from Microsoft's anti-spam message header documentation (updated 27 July 2026), which describes the Authentication-Results and X-Forefront-Antispam-Report headers in Microsoft 365 mail.
| Header value | A pass looks like | A problem looks like | What to do |
|---|---|---|---|
| spf= | spf=pass | softfail, fail or permerror | Add the sending tool to your one SPF record, then retest |
| dkim= | dkim=pass, with header.d= showing your domain | none (the message wasn't signed) or fail | Turn on DKIM signing in your mail provider or sending tool |
| dmarc= | dmarc=pass | fail; or bestguesspass or none, which both mean no DMARC record exists | Publish a DMARC record, starting with p=none |
| SFV: (Microsoft 365 only) | SFV:NSPM, marked as non-spam and sent to the recipients | SFV:SPM, marked as spam by spam filtering | Treat it as a failed placement and read CAT for the reason |
| CAT: (Microsoft 365 only) | No spam, bulk, phishing or spoofing category | SPM (spam), HSPM (high confidence spam), BULK, PHSH (phishing) or SPOOF (spoofing) | Fix authentication first, then look at volume and wording |
Gmail gives a quicker check. Under the sender's name, click the Down arrow: Gmail's authentication help page says a "Mailed by" line with your domain and a "Signed by" line with the sending domain mean the message is authenticated, and a question mark next to the sender's name means it isn't. The same page says Gmail treats messages signed with DKIM keys shorter than 1024 bits as unsigned, and warns against the DKIM length tag (l=).
Should you trust the SCL number in Microsoft headers?
Not on its own. Microsoft's spam confidence level page, updated 12 August 2026, says the SCL no longer means what it used to in cloud organisations: in Microsoft 365 it doesn't decide whether a message is treated as spam or what action is taken, and the same SCL value can appear on messages with different verdicts. Microsoft says to read CAT, which identifies what filtered the message, and DIR, which says whether the message was inbound, outbound or internal.
For a test sent to a Microsoft 365 mailbox, read these in order:
- SFV: NSPM means spam filtering marked it as non-spam; SPM means it was marked as spam.
- CAT: SPM, HSPM or BULK tells you which policy acted on it.
- SRV:BULK: the message was identified as bulk email. Microsoft says bulk mail is identified as spam when the MarkAsSpamBulkMail setting is on, which it is by default.
- compauth: Microsoft's composite authentication result, judged on the From domain. The documentation notes that a message can still be allowed despite a compauth failure, so treat a fail as a warning and check SFV and CAT as well.
How many results do you need before a test means something?
More than one. Placement percentages come from small counts, so do the arithmetic before you rewrite anything. The numbers below are a worked example, not a test we ran.
Say a seed test sends your email to 20 Gmail seed mailboxes and reports 16 in the inbox: 16 ÷ 20 = 80%. Each mailbox is worth 100 ÷ 20 = 5 percentage points, so a retest at 70% means 14 of 20, or two mailboxes moving. Rerun once before you change the copy. If two runs agree, act on it.
Seeds are also a particular audience. GlockApps' help centre (updated 7 October 2025) says its seed list "represents a cold audience where prior user engagement doesn't impact the email placement," with default filters, and calls its results accurate for that seed list. In our reading, that makes seeds a fair stand-in for a first cold email, but they can't tell you how a prospect's own filter rules or past behaviour will treat you.
Free test allowances cover a small team. Three mailboxes, each retested after two copy changes a week, is 3 × 2 = 6 tests a week, which fits inside mail-tester's 3 free tests per 24 hours if you spread them over the week. Testing 10 variants in one afternoon uses the 3 free tests and needs 7 more, and the smallest mail-tester pack is 10 tests for $5 before VAT, or $0.50 a test (checked 4 November 2026).
A perfect score isn't a promise either. Mail-tester's own FAQ says all spam filters work differently: "What may pass in mail-tester, might not in your own office inbox."
What should you test before every new campaign?
Run the same short test every time something changes, and write the result down so you can compare runs. Copy this into a sheet, one row per test:
Retest when you add a sending tool or edit your SPF record, start a new mailbox or domain, add a new link domain, or see replies drop on a sequence that used to work. If you change tracking settings, our post on whether open tracking hurts deliverability covers what the tracking pixel adds to each email. A brand-new mailbox should ramp slowly before it sends at full volume; see how long to warm up a new domain.
If Siela's Lead Finder sends your cold email, it goes from a Gmail mailbox you connect, so run the test from that same mailbox.
What can't a pre-send test tell you?
How a provider rates all of your mail over time. For mail to personal Gmail accounts (addresses ending in @gmail.com or @googlemail.com), Google's Postmaster Tools dashboards show spam rate, authentication, encryption, delivery errors and a Compliance status view. Google says the data is typically updated within 24 hours, may be incomplete on low-volume days, and that a fix can take up to 7 days to show in Compliance status.
Yahoo offers a Complaint Feedback Loop for DKIM-signed mail: its sender best practices page says that when users click report spam, you can get a copy of the complaint. You enrol through a Sender Hub account, according to Yahoo's sender FAQ.
When real sends start bouncing or being deferred, Google's sender guidelines say to reduce volume until the SMTP error rate falls, and if bounces and deferrals continue at a low volume, to review individual messages, for example by sending a blank test message. Our reading: if a blank message delivers and your campaign email doesn't, look at the content first; if both fail, look at the domain, IP or authentication. For what the paid testing and monitoring tools cost, see our deliverability tools price comparison.
Frequently asked questions
What is a good mail-tester score?
Mail-tester sets 10 out of 10 as the goal. It scores with a default SpamAssassin install and reverses the result, so a positive number is good and a negative one is bad. Its own FAQ warns that a perfect score can still land in spam, because every filter works differently, so treat 10 out of 10 as a clean setup check rather than a placement result.
How does the MxToolbox email deliverability test work?
You send an ordinary email to [email protected]. MxToolbox replies with a link to a full Deliverability Report, which it says analyses your message headers, the blacklist reputation of the IP address your mail left from and your SPF records. MxToolbox's page lists no price for it, and it pairs well with mail-tester because the two reports check different things.
Can you test email deliverability without sending an email?
Only partly. DNS lookups confirm that SPF, DKIM, DMARC and MX records exist and are formatted correctly, and a blocklist lookup checks your sending IP. Neither shows whether a real message passes authentication, how a filter scores the content or which folder it reaches. For that you need to send the actual email to a tester or to real inboxes.
How often should you run an email deliverability test?
Run one before each new campaign and again after any change to DNS records, sending tools, mailboxes, link domains or tracking settings. During a new mailbox's ramp, a weekly test per mailbox is a reasonable rhythm. Postmaster Tools data usually updates within 24 hours, so check it the day after a change rather than the same hour.
Why do deliverability tests give different results?
Each one tests something different. Mail-tester runs SpamAssassin, MxToolbox checks headers, SPF and blocklists, and GlockApps' seed tests log in to test mailboxes where past engagement doesn't affect placement. Gmail and Microsoft each judge mail with their own filters and signals. Compare runs of the same test over time, and trust two runs that agree over one dramatic number.
Sources
- mail-tester.com: Frequently Asked Questions (checked 4 Nov 2026)
- mail-tester.com: Pricing and account features (checked 4 Nov 2026)
- MxToolbox: Email Deliverability tool (checked 4 Nov 2026)
- Gmail Help: Trace an email with its full header (checked 4 Nov 2026)
- Gmail Help: Check if your Gmail message is authenticated (checked 4 Nov 2026)
- Google Admin Toolbox: Messageheader (checked 4 Nov 2026)
- Microsoft Support: View internet message headers in Outlook (checked 4 Nov 2026)
- Microsoft Learn: Anti-spam message headers in Microsoft 365 (updated 27 Jul 2026, checked 4 Nov 2026)
- Microsoft Learn: Spam confidence level (SCL) in Microsoft 365 (updated 12 Aug 2026, checked 4 Nov 2026)
- Microsoft: Message Header Analyzer (checked 4 Nov 2026)
- GlockApps Help: How accurate is the delivery report? (updated 7 Oct 2025, checked 4 Nov 2026)
- Gmail Help: Postmaster Tools dashboards (checked 4 Nov 2026)
- Yahoo Sender Hub: Sender best practices (checked 4 Nov 2026)
- Yahoo Sender Hub: FAQs (checked 4 Nov 2026)
- Google Workspace Admin Help: Email sender guidelines (checked 4 Nov 2026)
- Gmail Help: Set up Postmaster Tools (checked 4 Nov 2026)