Does One Bad Inbox Placement Test Mean Your Domain Is Burnt?

TL;DR: One bad inbox placement test does not mean your domain is burnt. A single low reading is usually noise, especially when it rests on a small number of seed inboxes, and a real burn only shows up as consecutive bad days that keep falling, confirmed by signals like a Gmail spam rate above 0.3% or inbox placement under 60%. Cedric Lauster treats a domain as burnt only when 2 of its last 3 test days come in under 85% with at least 10 Google seeds, or a single reading drops under 60% on at least 10 seeds.

Key Takeaways

• A single low placement reading is usually statistical noise rather than a sign of domain burn, especially on a small seed count.
• A genuine domain burn shows up as consecutive bad days that keep falling, not one isolated dip that recovers on the next test.
• Google's own sender guidelines put the spam-rate ceiling at 0.3%, measured over days, and a seed reading under 60% on at least 10 seeds is where Cedric treats a domain as burnt.
• Microsoft inboxes sit near 100% on most healthy fleets, so a blended placement score can hide a Gmail-specific problem.
• Causes of a bad placement reading unrelated to domain reputation include technical misconfiguration, poor list hygiene, authentication gaps, and provider mix.

No. A single low inbox placement reading is usually noise, especially when the percentage rests on a small number of seed inboxes. I only treat a domain as burnt when 2 of its last 3 test days come in under 85% on Google with at least 10 Google seeds across those days, or when a single reading drops under 60% on at least 10 seeds. Practitioners define a burnt domain with multiple signals crossing thresholds together, not with one bad test.

Look at what a burnt domain means in Google's own terms. Google's bulk sender guidelines tell senders to keep their spam rate below 0.1% and never let it reach 0.3%. That is a rate measured over days, not one test that looked bad.

If your reply rate dropped at the same time, the placement test is one input, not the diagnosis. I wrote a separate piece on how to diagnose a reply rate drop in the right order.

What is a seed-based inbox placement test, and why does the number move so much?

A seed test sends a copy of your campaign email to a panel of test inboxes (seeds) at Google and Microsoft, then reports how many landed in the inbox versus spam. It is the closest thing most senders have to seeing where their mail actually lands, which is why the number carries so much weight and gets over-read so often.

The reason it moves is sample size. A daily test usually sends somewhere around a dozen to twenty seeds per domain to Google. On 10 seeds, one email in spam moves the headline percentage by 10 points. That is not a reputation change. That is arithmetic.

How many seeds are actually behind the percentage?

I never read a placement percentage without the seed count next to it. A clean-looking number on a handful of seeds tells you almost nothing, and a test can report a tidy percentage when many of the seeds it tried to reach never showed up at all.

So two questions, every time:

• How many Google seeds does this percentage rest on?
• How many of the seeds the test sent actually arrived?

If the answer to the first question is under 10, the number is directional at best. I run placement tests every weekday on the mailboxes that are actually sending, and I read the Google result, not the blended one. Microsoft seeds sit near 100% on almost every fleet I have measured, so a blended score quietly hides a Gmail problem behind a Microsoft pass.

What does a genuine domain burn actually look like over several days?

Real trouble shows up as consecutive bad days that keep falling. A single dip that recovers on the next test is not a burn, it is variance.

Here is what that looked like across my own fleet in daily Google-seed tests from mid September to early October 2026. Two domains were genuinely in trouble, and both showed it on consecutive days:

• Domain A: 80% (16 of 20 seeds in the inbox), then 76% (13 of 17), then 64% (9 of 14).
• Domain B: 79% (15 of 19), then 80% (16 of 20), then 69% (9 of 13).

Now compare that with a domain that dipped once. It read 80% on a single test, which was 10 Google seeds with 2 in spam. Its next test read 100% (15 of 15). If I had pulled that domain on the first reading, I would have removed a perfectly healthy sender from the rotation.

The healthy domains in the same window were not flat either. They posted occasional small dips, 90%, 91%, 92%, 94%, and went straight back to 100% on the following tests. Small, isolated, self-correcting.

One caveat, stated plainly: this is one fleet over a few weeks with two genuine failures in it. It shows a pattern worth watching. It is not a statistical threshold, and I would not present it as one.

In my experience, the biggest cost of misreading a seed test is not the burnt domain you keep sending from. It is the healthy domains you retire out of nervousness, one by one, until your sending capacity is a fraction of what you paid for and you still have not found the real problem. Direction over days beats precision on any single day.

What causes a bad placement reading that has nothing to do with domain reputation?

Plenty. A bad reading is a symptom, and the domain's reputation is only one of several things that can produce it.

Technical misconfiguration. A template change, a tracking parameter, a new signature block or one broken header can tank a test on a domain with a spotless history.

List hygiene. Invalid and risky addresses inflate bounce and complaint signals, and in a single snapshot that looks identical to a reputation problem. Worth noting that over-aggressive verification causes its own problem: see why email verifiers can delete real buyers on catch-all domains.

Authentication gaps. If SPF, DKIM or DMARC broke on one domain after a DNS change, the placement test shows you the effect without telling you the cause.

Provider mix. A blended score mixes Google and Microsoft seeds. On almost every fleet I have measured, Microsoft seeds sit near 100%, so a blended number can look healthy while Gmail is failing. Read the Google result on its own.

If you want a second opinion on your own infrastructure before you change anything, that is a reasonable thing to get from someone who runs this daily rather than from a forum thread.

What should you check before you pause or change a domain?

Check the signals that update independently of the seed test, over a window long enough to show a trend.

1. The Google-only result. Not the blended score. Microsoft seeds sit near 100% on most healthy fleets, which means a blended number can look acceptable while Gmail placement is sliding.

2. Google Postmaster Tools spam rate, over at least 7 days. Google's bulk sender guidelines tell senders to keep their spam rate below 0.1% and to prevent it from ever reaching 0.3% or higher. Google also describes enforcement as graduated and threshold-based rather than a single-event judgment: delivery mitigation eligibility is pulled once the spam rate exceeds 0.3%, and restored after it stays below that for a sustained period. That is the same logic you should apply to your own reading of the data.

3. A blacklist check and your bounce and complaint trends. Mailpool recommends starting diagnosis with a blacklist check and a review of deliverability metrics rather than assuming the worst from one signal. ZeroBounce (2025) notes that a bounce rate higher than 2% negatively impacts sender reputation, which makes bounce rate a reputation problem on its own, with or without a seed test.

4. The trend, not the snapshot. Mailbeast (2026) recommends judging reputation recovery or decline over a multi-week trend rather than any single daily reading, since Postmaster Tools recalculates reputation daily.

If the reply rate is what sent you looking at placement in the first place, work through the full diagnosis order for a reply rate drop instead of fixating on one tool.

What is the diagnostic to run the next time a test comes back low?

Look at two things before touching anything: how many Google seeds the percentage rests on, and what the same domain read on its last two tests.

Then act on the pattern:

• One low day, small seed count. Retest tomorrow. Change nothing. This is a watch, not an incident.
• Second or third low day in a row, with at least 10 Google seeds across them. Act. Pull the domain out of the campaign and diagnose.
• A reading under 60% on at least 10 seeds. Treat it as burnt and move to rehabilitation rather than retesting hopefully.

If a domain really is burnt, the repair is unglamorous: stop cold sends from it, fix SPF, DKIM and DMARC alignment, clean the list, and re-warm it gradually before it carries campaign mail again.

One supporting view on how much weight to give the test itself: Eric Nowoslawski, founder of Growth Engine X, laid out a five-level deliverability diagnostic in a YouTube video on 30 August 2026. Daily seed tests sit at level two, and he treats them as directional only, a signal to watch rather than a verdict on their own. His other levels include warmup reputation, weekly reply rate per inbox, and a test against auto-responder addresses.

While you are rebuilding sending capacity, it is worth re-checking your volume assumptions too: here is how many cold emails one mailbox can safely send and what counts as a good cold email reply rate once mail is landing again.

When is it not worth trying to save the domain?

When the signals agree. Repeated readings under 60% on at least 10 seeds, a Gmail spam rate above the 0.3% line Google names, and a "Bad" Postmaster reputation together describe a domain that is burnt, not a domain having a bad week. At that point, retesting indefinitely is just delay.

The decision deserves to be deliberate, because the recovery window is measured in weeks or months depending on severity. That is real sending capacity out of rotation. Making that call off one bad day is how founders end up with a graveyard of retired domains and a pipeline that never got the chance to build momentum.

The better move is upstream: authenticate every sending domain properly, verify every address before it goes out, and keep spam complaints under the 0.1% Google asks for. Most of the "is my domain burnt" panic I see traces back to one of those slipping, not to anything Gmail decided about the domain.

What is the next step if your placement numbers are unstable?

If your seed tests are swinging and you cannot tell whether it is infrastructure, list quality, or a broken template, the fastest way through is to look at the whole sending setup at once instead of reacting to one reading at a time. Book a Growth Mapping Call and I will walk through your sending infrastructure, your list, and your authentication with you, and tell you plainly which part is actually causing the problem.

Frequently Asked Questions

How many Google seeds should a placement test include before I trust the percentage?

Cedric Lauster treats anything under 10 Google seeds as directional at best, since on 10 seeds one email landing in spam moves the headline percentage by 10 points. He reads the Google-only result, not a blended score, because Microsoft seeds sit near 100% on most healthy fleets and can mask a Gmail problem.

What percentage counts as a burnt domain?

Cedric Lauster treats a domain as burnt when a seed reading falls under 60% on at least 10 Google seeds, and takes it out of campaigns when 2 of its last 3 test days come in under 85%. Google's bulk sender guidelines add the other hard line: a spam rate that reaches 0.3%.

Why does list quality affect inbox placement tests?

Invalid and risky addresses inflate bounce and complaint signals, which in a single snapshot looks identical to a reputation problem. Verify every address before it is sent, so a list problem is not misread as a domain problem.

Does authentication status change inbox placement results?

Yes. A DNS change that breaks SPF, DKIM or DMARC on one domain can tank a placement test without any real reputation damage, so check authentication before concluding the domain is burnt.

How long does it take to recover a burnt domain?

There is no reliable universal timeline, it depends on how much damage was done. Cedric Lauster takes the domain out of campaigns, fixes the cause, keeps it warming, and only puts it back once its tests hold up again on consecutive days.

Should I check Google Postmaster Tools before pausing a domain?

Yes. Google's bulk sender guidelines recommend keeping spam rate below 0.1% and never letting it reach 0.3% or higher, with enforcement applied over a sustained period rather than a single event. Cedric Lauster recommends checking this over at least 7 days before acting on one seed test.

What should I check before pulling a domain out of rotation?

Check the Google-only placement result, the Google Postmaster Tools spam rate over at least 7 days, a blacklist check, and bounce and complaint trends. Mailpool recommends starting with a blacklist check and reviewing deliverability metrics rather than assuming the worst from one signal.