How should results be interpreted (directional, not absolute)?

Seedlist results are directional, not absolute. Treat them as a weather report, not GPS coordinates.

Here is what that means in practice. A seedlist is a panel of fake mailboxes you control across Gmail, Yahoo, Outlook, Apple, and a long tail of regional providers. Your ESP or a tool like Everest, GlockApps, or Inbox Monster sends your campaign to those seeds, then reports where each one landed. You get a row per provider that says inbox, spam, tab, or missing. For background on the mechanics, see how seedlist testing works.

The trap is treating that row like a percentage of your real audience. It is not.

What seed results actually tell you

• General placement health at each provider. If Gmail seeds all landed in spam, something is wrong with how Gmail sees you right now.
• Authentication problems. Missing SPF alignment, broken DKIM, DMARC failures show up consistently across seeds because seeds get the same headers your real subscribers do.
• Content red flags. If only the Outlook seeds tank, the content tripped a Microsoft filter, not a reputation issue.
• Provider-specific behavior. Gmail might inbox while Yahoo spams the exact same send. That gap is real and worth digging into.

What they cannot tell you

• Exact placement for your actual list. Seed mailboxes have no open history, no click history, no "move to inbox" votes, no contacts list entry for your sending domain. Real subscribers do.
• User-level variation. Two Gmail users on the same campaign can get different placements because Gmail personalizes filtering per recipient. More on that in why identical campaigns land in inbox for some and spam for others.
• Engagement-influenced outcomes. Your most engaged subscribers almost always inbox better than seeds. Your dormant ones almost always inbox worse. Google has been explicit that user engagement signals drive placement decisions (Gmail sender guidelines).

How to read the numbers

If your seeds show 80% inbox at Gmail, do not tell the team "80% of subscribers got the inbox." You do not know that. What you do know is that a clean panel with no engagement history got inbox 80% of the time. Your engaged segment is probably doing better. Your six-month dormant segment is probably doing worse. The real spread might be 95% inbox for openers and 40% for the dead weight, averaging out to something close to 80, or nowhere near it.

This is why seeds and your ESP "delivered" number disagree, and why both disagree with what subscribers actually see. If that disconnect is news, read what "delivered" actually means in ESP reports and how to measure inbox placement accurately.

When to act on a seed result

Use seeds as a trigger, not a verdict.

• A Gmail seed in spam warrants investigation even if you cannot prove the real-world impact. Check Postmaster Tools, look at domain reputation, pull the spam-flagged seed's headers and compare to an inboxed seed.
• A sudden drop across multiple providers usually means an authentication break, a blocklist hit, or a content change. Run it down the same day.
• A single seed missing at one provider is noise. Do not rebuild your program over it.
• Trend over time matters more than any single send. One report is a snapshot. Ten reports across two weeks is a pattern.

The rule we use at Review My Emails: pair seed data with Google Postmaster Tools, Microsoft SNDS, your bounce and complaint rates, and the engagement metrics from your ESP. Seeds tell you what the filters are doing in a vacuum. The other signals tell you what your audience is doing. You need both.

If the seeds and the engagement data disagree, trust engagement. Real subscribers opening your mail is the only metric the filters care about long-term.