Inbox placement testing shows what happened after the receiving mail server accepted the message.1 Delivery tells you a receiving server accepted the email.2 Placement tells you where the mailbox put it.3 Use a sample message containing the content planned for customers.4 Send it to controlled accounts across providers.5 A delivered status does not prove an inbox arrival. A platform can report delivery while the message sits in Promotions, Junk, or a rarely checked folder.6 When a reply-rate collapse points to domain reputation, check placement before retiring the sequence.7
Run the test
Start with a clear signal, use a representative message, and set a stopping point. Work through the stages in order so you can separate acceptance, folder placement, provider behavior, and message content.
| Stage | What you are trying to learn | Example question |
|---|---|---|
| Trigger | whether the signal warrants a placement test | What changed in the sending results? |
| Prepare | whether the sample matches the planned send | Is this the message customers will receive? |
| Send | whether controlled accounts accept the message | Did each account receive a copy? |
| Classify | where each copy landed | Did it reach the inbox, another folder, or nowhere? |
| Diagnose | which provider or message condition needs work | What do the headers and complaint data suggest? |
| Retest | whether the change affected placement | Did the same message improve after the change? |
Start with the signal
Choose what to compare based on why you are running the test. A placement check is most useful when you can connect its result to a change in replies, opens, complaints, or bounces.
Monitoring inbox placement rate gives you a more accurate view of the sending program's health.8 Compare bounce and complaint rates at the provider level, then look for drops in open or click rates that point to bulk delivery to spam at particular providers.9 Move on when you know which provider pattern you are trying to explain.
Prepare the sample
Use a copy of the customer message. Keep its content and sending mechanics close to the planned send so the test shows the path of the live message.
A test message may be generated outside the normal campaign sending pipeline.10 Keep the sample on the live sending path where practical, and match its subject, body, sender identity, links, and tracking to the planned message. Avoid sensitive, confidential, or personally identifiable information because real data is unnecessary for the test.11 Move on when the sample reflects the message you intend to send and contains safe test data.
Send to controlled accounts
Keep the test audience separate from real recipients. That gives you a repeatable view of provider behavior without interrupting campaign recipients.
A seed test uses controlled provider accounts so results can be compared without bothering real subscribers.12 Coverage can extend across more than 95 major email providers in a placement test.13 Choose the providers that matter to the audience you are investigating, and keep the sample unchanged across them. A test completes after about 24 hours, after which its results can be viewed.14 Wait for the test window to finish before deciding that a provider accepted, filtered, or missed the message.
Read the result
Read the output in two steps: did the message arrive, and where did the mailbox put it? Record the outcome by provider so a broad average does not hide a provider-specific failure.
Test results show how many messages reached test recipients' inboxes.15 They also show how many went to junk folders and how many were not delivered.16 Record those outcomes separately. A delivered result with a junk placement needs a different response from a missing message.
Check the message headers, authentication results, and placement together.17 For mail accepted but placed in spam, capture provider-specific placement tests, complaints, Postmaster data, and message headers.18 A spam score points to possible content, authentication, or sender problems.19 Use that score to choose the next inspection, then confirm the suspected issue with the provider-level result.
Move on when each provider has a clear outcome and you have a plausible cause to test. A vague result such as poor deliverability leaves too many possible fixes in play.
Fix, then retest
Match the change to the failure you found, then run the same sample through the same provider set. Changing several conditions at once makes the next result harder to interpret.
Placement tests can identify deliverability problems caused by email content.20 Test across mailbox providers, fix the conditions affecting placement, give recipients instructions for adding the sender address to their address book, and, if needed, change the From email and From name after notifying subscribers.21 Retest before deciding whether the sending sequence itself needs to change.
What not to do
Shortcuts can answer a neighboring question and leave placement unresolved.
- Do not use a platform's Send Test feature to evaluate inbox placement.22
- Do not treat a result from your own Gmail or work account as a forecast for the campaign.23
- Do not let the test message differ materially from the message subscribers receive.24
- Do not wait until open rates tank after messages have landed in spam for 500 sends.25