If you're searching for how to check inbox placement for cold emails, most guides stop at a seed test and a spam score. That covers half the job. This page gives you a test log you can copy. It is weighted toward the work mailboxes your prospects actually read, and it ends in one decision: send, fix, or pause the domain.
How do you check inbox placement for cold emails?
Send your real campaign email from your real mailbox to a seed list of test inboxes, then record where each one landed: inbox, other tab, spam or missing. Then confirm the result with live signals: bounce codes, your Gmail Postmaster spam rate, and reply rate by sending domain.
This matters more now than it did two years ago. Starting November 2025, Gmail began ramping up enforcement on non-compliant traffic. Mail that fails its sender requirements can get temporary or permanent rejections instead of a quiet trip to spam [5].
The template below puts the whole process on one screen.
Cold email inbox placement test template (copy this)
PLACEMENT TEST: [campaign] / [sending domain] / [date]
1. Sending mailbox: ______@______ (the real one, not a test account)
2. Message: exact campaign email
same subject / same links / same signature
no "test" in the subject
3. Seed mix (total __), work mailboxes = majority:
Google Workspace ......... __
Microsoft 365 ............ __
Corporate gateway filter . __
Personal Gmail ........... __
Outlook.com .............. __
4. Results by provider: Inbox | Other tab | Spam | Missing
Google Workspace __ | __ | __ | __
Microsoft 365 __ | __ | __ | __
Corporate gateway __ | __ | __ | __
Personal Gmail __ | __ | __ | __
Outlook.com __ | __ | __ | __
5. Spam score: ____ (record it; do not pass on it alone)
SPF __ DKIM __ DMARC __ Blacklists: __
6. Bounces (last 7 days): code ______ count __
4.7.x (temporary) __ 5.7.x (permanent) __
7. Postmaster v2 spam rate: ___% (vs 0.1% and 0.3%)
8. Reply rate by sending domain:
last 7 days ___% vs prior baseline ___%
9. Decision: [ ] Send [ ] Fix and retest [ ] Pause domain
Reason: ______________________
Here is what to put on each line:
- Sending mailbox. Send the test from the exact inbox you plan to use for the campaign [2]. A fresh test account has a different reputation and tells you nothing about your real one.
- Message. Paste the real email: same subject, links, tracking and signature. A plain-text test version can inbox while the real one, with its links, gets filtered.
- Seed mix. Write down counts per provider so you can weight the results later. Work mailboxes should be most of the list (more on this below).
- Results by provider. Count each seed as one outcome. "Missing" counts as a failure. It often means a gateway or rejection dropped the message.
- Spam score. Record the SpamAssassin score along with the SPF, DKIM and DMARC results and any blacklist hits [1][2][3]. A failed authentication check is a fix-first item. A low score is not a pass.
- Bounces. Log the exact code and the count. A 4.7.x and a 5.7.x lead to different actions.
- Postmaster v2 spam rate. Google recommends staying under 0.1%. 0.3% is the policy-violation line [10].
- Reply rate by sending domain. Compare the last 7 days with the domain's own history, not with an industry average.
- Decision. Tick one box and write one line explaining why. Next month you will want to know what you saw.
For the message line, use the email you will actually send. Here is an illustrative one built to the 4-line pattern from our short B2B cold email template. The numbers are placeholders, so swap in your own real figures:
Subject: carrier invoice reconciliation
Hi [First name],
I ran billing ops at a 3PL for six years before building [Product].
It's for finance leads at freight brokers who still match carrier
invoices by hand. Teams using it cut month-end reconciliation from
[3 days] to [1].
Is reconciliation still a manual job on your team?
[Your name]
Which seed inboxes should B2B cold emails be tested against?
Most default seed lists lean toward consumer inboxes. Your prospects mostly read Google Workspace and Microsoft 365 accounts, often behind a corporate security gateway. A seed list weighted toward those professional inboxes reflects your real audience more closely [15].
Here are your main tool choices:
- GlockApps sends your actual email to 70 seed addresses across mailbox providers and runs it through 5 spam filters [14]. That gives you a per-provider split, which is what the results table needs.
- Mail-tester does not show where your email lands across providers, and the free version allows 3 tests a day [2]. Use it to check content and authentication, not placement.
- Built-in tools in your sending platform can work too. Check which providers their seeds cover before you trust the result [1][3].
One practical step: if your list is almost all company domains, drop or down-weight the personal Gmail and Outlook.com rows when you calculate your placement number. Inboxing to a personal Hotmail account doesn't help you reach a VP at a 300-person software company.
Why a good spam score doesn't mean you reached the inbox
This is the most common false pass. In Unspam's 2026 report, 88% of tested emails scored under 3.0 on SpamAssassin, but only 66% reached the inbox or a tab [12]. That leaves a 22-point gap between "looks clean" and "arrived."
Treat that figure with some care. It is a vendor dataset drawn from tests run through Unspam, and the published summary gives no total test count [12]. The direction still holds: content scoring is one input among many that mailbox providers weigh.
Seed tests have their own limit. Seed inboxes are probes, not people. They have no reading history with you and no relationship with your domain. GlockApps itself says a seed test does not tell you what share of your whole list lands in spam [14][15].
So a seed test is a hypothesis. The next three signals tell you whether it held up with real recipients.
How to tell if cold emails are going to spam using bounces, Postmaster and replies
1. Bounce codes are placement data now. Since November 2025, Gmail has issued temporary 4.7.x and permanent 5.7.x rejections to non-compliant mail [5][6]. Microsoft rejects non-compliant bulk mail with 550 5.7.515. Some sources write it as 5.7.15 [7][8]. A 4.7.x means fix the cause and let the retry happen. A 5.7.x means stop and fix before you send to that provider again. A cluster of either in your log beats any seed test as evidence.
2. Postmaster v2 spam rate, read carefully. It counts only Gmail messages that were delivered to the inbox and then manually marked as spam [11]. If Gmail is filtering you, fewer people see you and fewer can complain. A 0.0% reading combined with a drop in volume can therefore mean filtering, not good health. Run a placement test when you see that pattern [11]. If you cross 0.3%, you can lose Gmail's mitigation support until you have been back under 0.3% for seven consecutive days [10].
3. Reply rate by sending domain. Break replies out per domain so one bad domain can't hide inside a healthy average. A practitioner rule of thumb: a sudden drop in reply rate, especially below 1% on a campaign that was performing before, almost always means placement loss [4]. Before you blame placement, rule out a copy change. A weak ask can also tank replies, which is why it's worth reviewing your call to action.
Small senders are not exempt. Google counts you as a bulk sender at close to 5,000 messages a day to personal Gmail accounts, and that status does not expire once assigned. Its guidelines apply to every sender regardless of volume [16].
What inbox placement rate is good for cold email?
One cold email guide sets the target at 85% or higher [2]. For context, Validity's 2026 benchmark puts global average inbox placement at 87.2% in 2025, with Microsoft/Outlook at 77.4%. That data covers opted-in marketing mail, not cold email, and we are citing it secondhand [13]. Unspam's 2026 vendor data shows primary inbox placement at 63% and spam at 33% across the tests it ran [12].
Our editorial rule of thumb for the Decision line:
- Send if your work-mailbox seeds clear the target and bounces, Postmaster and reply rate are all clean.
- Fix and retest if the seeds are weak but the live signals look fine.
- Pause the domain if the seeds show spam or missing, and reply rate or bounces confirm it.
What to do next if your cold emails are landing in spam
Fix in this order [1][4]:
- Authentication. Make sure SPF, DKIM and DMARC all pass and align with your sending domain.
- List cleaning. Verify addresses and remove bounces and role accounts.
- Content. Cut links and tracking, and shorten the email. Our guide on how to write a cold email as a founder keeps it lean.
- Retest with the same template so the before and after are comparable.
Retest after every fix and before every new campaign or new domain.
If you'd rather not run this test by hand each month, Outbound by ReachSmart.ai measures inbox placement and spam signals on every campaign. It sends from your own Gmail or Microsoft mailbox and learns from every reply. Pricing is per prospect with no seats or contract, and you can see the number for your volume on the calculator at https://outbound.reachsmart.ai. Start here: https://outbound.reachsmart.ai/signup