Inbox placement rate is the share of delivered emails that actually reach the inbox, rather than the spam folder or a promotions tab. Seed testing measures it by sending a sample of every campaign to a panel of test inboxes across Gmail, Outlook, Yahoo and others, so you can see placement before your real audience does.
What inbox placement rate means
Of the messages that were delivered (accepted by the receiving server), what percentage landed in the primary inbox rather than spam, junk or a promotions tab. A delivery rate of 99% paired with a 60% inbox placement rate means a third of your accepted mail is invisible to recipients.
How seed testing works
You maintain (or buy access to) a controlled set of test inboxes across the major mailbox providers. Each campaign is sent to those seed addresses alongside the real audience. A monitoring service then reports where the message landed in each test inbox, giving you a per-provider placement breakdown.
What seed tests can and cannot tell you
Seed tests are a fast, cheap way to spot obvious problems: a clear pattern of spam folder placement at Gmail, for example, or a Microsoft block. They cannot perfectly predict placement for your real recipients, because providers weight engagement signals from each individual address.
Treat seed tests as a directional smoke test, not the gospel. Always cross-reference with real engagement: opens, clicks and replies from the actual send.
How to act on the results
- If one provider is consistently bad, focus there: check reputation tools, review complaints, audit authentication.
- If placement drops after a content change, revert and isolate the variable.
- If placement is poor across the board, the problem is reputation or list, not content.
Frequently asked questions
What is the difference between delivery rate and inbox placement rate?
Delivery means the server accepted the message. Inbox placement means a human is likely to see it. The first is a vendor success metric; the second is the operator one.
How big does a seed list need to be?
A few hundred seeds across the main providers is enough for directional reads. Larger seed panels improve confidence but add diminishing value.
Can I run seed tests on every send?
Yes, and at scale you should. The marginal cost is tiny and a placement drop on a single send is often the earliest warning of a reputation problem.