A 99% delivery rate is compatible with nobody ever seeing your email. Delivery means the receiving server accepted the message. It says nothing about whether the message went to the inbox, to promotions, or straight to spam, and the difference between those three outcomes is the difference between a working campaign and a silent one.
An inbox placement test is the only way to see which one you are getting, and it is the diagnostic most deliverability checklists leave out.
What the test does
You send your real campaign email to a seed list of monitored mailboxes across the major providers. Each seed reports which folder the message landed in. The output is a placement breakdown per provider rather than a single number.
That per-provider split is the whole value. Gmail, Microsoft and Yahoo filter on different signals, and a domain that lands cleanly at Gmail while going to junk at Outlook has a specific, fixable problem, not a general reputation problem.
Running it so the result means something
- Test the real email. Same copy, same links, same signature, same tracking domain. A stripped-down test message tells you about a message you are not sending.
- Test per sending domain and per mailbox group. Reputation is per domain and largely per IP, so a single test on one mailbox does not describe a 40-mailbox setup. Test a sample from each domain.
- Send from the warmed sending profile, at the pacing you actually use. A test blasted from a cold mailbox measures the test, not the campaign.
- Do not send only to seeds. Seeds never open or reply, so a mailbox whose only traffic is seed tests will look inert. Run tests alongside real sending.
- Baseline before you need it. Test when things are healthy so you know what normal looks like for your setup. Placement data with no baseline is hard to act on.
Reading the result
| Pattern | Most likely cause | First move |
|---|---|---|
| Inbox at Gmail, junk at Microsoft | Microsoft weighs domain age and volume consistency harder | Slow the ramp on those domains, keep volume steady day to day |
| Promotions tab at Gmail | Message looks commercial: links, images, marketing phrasing | One link maximum, no images, plain signature |
| Spam everywhere, one domain | That domain or its IP has a reputation problem | Stop sending on it, check blacklists, do not rotate the problem onto a fresh domain |
| Spam everywhere, all domains | Content, list quality, or the tracking domain | Check the list first, bounces from a stale list cause this more often than copy does |
| Degrading week over week | Volume ramped faster than reputation | Cut volume by half and hold for two weeks |
One caveat worth keeping in mind: seed results are a strong indicator, not a guarantee. Real recipients open, reply and mark things important, and that engagement moves placement in ways a seed list cannot reproduce. Read placement tests as a directional diagnostic, and read reply rate as the truth.
How often
- Before any new domain sends to real prospects. Non-negotiable, and it is what test allowances are for.
- Weekly during a ramp, while volume is changing.
- Monthly at steady state, per domain.
- Immediately when reply rate drops without a copy change. This is the alarm that placement testing exists to explain, and it is usually placement rather than persuasion.
Where it fits in the diagnostic order
Placement testing is step four, not step one. Fix SPF, DKIM and DMARC first, then warm the mailboxes, then verify the list, and only then test placement. Running a placement test on an unauthenticated domain tells you what you already know.
Next: the full deliverability guide, or why cold emails go to spam, in the order the causes actually occur.
Frequently asked questions
What is an inbox placement test?
A send to a seed list of monitored mailboxes across the major providers, which reports which folder each message landed in. It measures placement, inbox, promotions or spam, rather than delivery.
Why is delivery rate not enough?
Delivery only means the receiving server accepted the message. A 99% delivery rate is entirely compatible with every message going to spam.
How often should I test inbox placement?
Before any new domain sends to real prospects, weekly while volume is ramping, monthly per domain at steady state, and immediately if reply rate falls without a copy change.
What does landing in the Gmail promotions tab mean?
The message reads as commercial. Reduce to one link, remove images, and simplify the signature. It is a content signal rather than a reputation problem.
Are seed list results reliable?
Directionally, yes. Seeds never open or reply, so they cannot reproduce the engagement that moves real placement. Treat placement tests as a diagnostic and reply rate as the truth.