Comparisons

Inbox placement tests vs Postmaster vs your own data

Compare seed-list inbox placement tests, Google Postmaster Tools and your own bounce and reply data for cold email, with a simple routine for small senders.

Summary

Three checks get called "deliverability monitoring." They answer three different questions.

  • Seed-list placement test: where did a test message land?
  • Google Postmaster Tools: what spam rate does Google measure for your sending?
  • Your own reply and bounce data: what did your real recipients and their mail servers do with your real emails?

Only the third covers the people you actually wrote to. The second is a reading from the provider itself, but only from one provider. The first is a controlled experiment on addresses that are not your prospects.

A small sender does not need all three at the same weight. Real send data comes first, because it is always there and it is about your own list. Postmaster Tools is a good second check on Google's view. A seed test is an occasional extra after you change something.

Differences

Seed-list placement testGoogle Postmaster ToolsYour own bounce, reply and unsubscribe data
What it measuresWhere a test message landsThe spam rate Google measures for your sendingWhat happened to the emails you actually sent
Whose viewTest addressesThe mailbox providerYour recipients and their servers
Covers your real listNoOnly the part that goes to GoogleYes
Main way it misleadsTest addresses are not your prospectsSays nothing about other providersSmall volumes make rates jump

The sections below take each one in turn.

Seed-list placement tests

A seed-list test is not a reading from a mailbox provider, and it is not data from your own sends. Its addresses are not your prospects, so treat a result as a proxy for your recipients, not a measurement of them.

When to look. Use it as a one-off comparison after you change something, such as the domain, a mailbox, the copy or a link. It is not a standing dashboard.

For a placement-test tool, see GlockApps pricing, plans and alternatives.

Google Postmaster Tools

Google's sender guidelines say: "Keep spam rates reported in Postmaster Tools below 0.3%" (Google's email sender guidelines, checked 2026-10-22) [[1]](#sources).

That line tells you two things. Postmaster Tools reports a spam rate that Google measures for your sending, and Google publishes a number to stay under [[1]](#sources). Unlike a seed test, this is the provider's own reading. Unlike your send logs, it includes a signal you cannot see yourself: how recipients on Google's side flagged your mail as spam.

Where it misleads.

  • It is one provider's view. Mailbox providers other than Google set their own rules, and Postmaster Tools does not speak for them [[2]](#sources) [[3]](#sources).
  • It is a rate, not a cause. A rising spam rate tells you something is wrong. It does not tell you whether the cause is the list, the copy, the sending volume or a mailbox that has not warmed up. For that you go back to your own send data.
  • It reflects what Google measures. A good reading is not proof that your emails reach the inbox in every case.

When to look. Check it when a campaign is live and whenever your own data turns. The point is to compare the spam rate against the 0.3% line [[1]](#sources). If it climbs toward that line, pause before it crosses.

Your own bounce, reply and unsubscribe data

This is the data that comes from real sends. In Outlevo's Smartlead integration, Smartlead sends Outlevo these events by webhook: sent, reply, bounce, unsubscribe and category changes [[4]](#sources). Outlevo builds each campaign in Smartlead, Smartlead sends it from the mailboxes in your Smartlead account, and Outlevo reads replies and bounces back from Smartlead by webhook [[4]](#sources).

Each event answers something different. The readings below are our own rules of thumb.

  • Bounces show messages that did not get delivered. When they rise, look first at the quality of your list and at your domain setup.
  • Replies show that a person read and answered. They are the closest sign in your data that mail reached a human.
  • Unsubscribes show that a recipient saw the email and did not want it.
  • Category changes arrive as events too, alongside the others.

The event list above has no complaint event [[4]](#sources). For complaints, the spam rate in Postmaster Tools is the place to look on Google's side [[1]](#sources).

Reference points for reply data. Smartlead's State of Cold Email 2026 puts the median cold email reply rate at 0.74% per contact. That is replies divided by people contacted, so a person who replies once counts once [[5]](#sources). The same report, covering more than 850 million emails sent from January to June 2026, puts the top quarter of senders at 1.43% and the top tenth at 2.63% [[6]](#sources).

Use those numbers as a rough ruler, not a target. A reply rate mixes deliverability with the offer, the list and the copy. A low rate can mean mail is landing in spam, or it can mean the message is wrong for the people it reached.

Where it misleads.

  • Small volume makes rates noisy. With a small list, one extra reply moves the rate a lot. Read trends across several sends instead of one day. How to A/B test cold email at low weekly volume covers working with small numbers.
  • Silence is ambiguous. No reply could be spam placement, no interest or a busy prospect. Your own data alone does not separate them.
  • Bounces are not the whole story. An email can be accepted by the server and still land in a spam folder. Acceptance does not equal inbox placement.

When to look. Continuously, because the events arrive as the emails go out. A bounce spike on the first batch matters more than any weekly report.

Before any of the three: authentication

None of these checks helps if the basics are broken. Three outside sources are worth knowing, each checked on 2026-10-22.

  • Microsoft announced for Outlook: "For domains sending over 5,000 emails per day, Outlook will soon require compliance with SPF, DKIM, DMARC." [[2]](#sources)
  • Yahoo's sender best practices: "Publish a valid DMARC policy with at least p=none" [[3]](#sources)
  • RFC 7208, the SPF standard, says SPF implementations "MUST limit the total number of those terms to 10 during SPF evaluation" [[7]](#sources). An SPF record that exceeds this can fail even when it looks correct [[7]](#sources).

For a DMARC tool, see dmarcian pricing, plans and alternatives.

Outlevo builds a check into sending. It will not send a campaign through Smartlead until every mailbox is connected and warmed up, and its domain has SPF, DKIM and DMARC checked within the last 7 days [[4]](#sources). Details are on the Smartlead integration page.

If you use Outlevo's managed mailboxes, a new mailbox sends nothing until it is warm: at least 14 days, usually about 3 weeks [[8]](#sources). Outlevo checks each mailbox's warm-up state and reputation daily [[8]](#sources). See the managed mailboxes page.

When to choose which

Choose your own send data when you want to know what is happening to the list you are actually emailing. It is the only one of the three that covers your real recipients, and it is the one you already have. Start here.

Choose Postmaster Tools when a meaningful share of your list is on Google-hosted mail and you want the provider's own spam reading to compare against 0.3% [[1]](#sources). It adds the one signal your logs lack.

Choose a seed-list test when you changed something and want a controlled comparison without risking real prospects. Examples are a new sending domain, a new mailbox, rewritten copy or a different link. Treat the result as a proxy for your recipients, not a measurement of them.

Do not use any of them as the only check. A clean seed result, a quiet Postmaster dashboard and a low bounce count can each hide a problem the others would show.

A simple routine for a small sender

This routine is built around events rather than a calendar, so you can keep it with a few mailboxes and one person.

  1. Confirm authentication first. SPF, DKIM and DMARC are in place, and the SPF record stays within the 10-term limit [[7]](#sources). Outlook's announced requirement and Yahoo's best practice set the bar above [[2]](#sources) [[3]](#sources).
  2. Let mailboxes warm up. Send nothing from a new mailbox until it is warm [[8]](#sources).
  3. Run one seed test before the first real send, and again after any change to the domain, mailbox, copy or links.
  4. Write down your stop line before you send. Pick the bounce and unsubscribe levels at which you will pause. Decide this in advance, so you are not arguing with yourself after a bad day.
  5. Watch the events as they arrive. Read bounces, replies, unsubscribes and category changes from the first batch [[4]](#sources).
  6. Check Postmaster Tools for the spam rate against Google's 0.3% line while a campaign is live [[1]](#sources).
  7. Compare reply rate over several sends against the reference points, 0.74% median, 1.43% top quarter and 2.63% top tenth, per contact [[5]](#sources) [[6]](#sources).
  8. Pause when a signal turns bad. Outlevo's kill switches pause a campaign, a channel or the whole workspace at once [[9]](#sources). Fix the cause first: list, copy, authentication or mailbox. Then resume.

Review on a schedule you can keep. A short check you actually do beats a thorough one you skip.

What Outlevo does here

Outlevo reads the events Smartlead returns and runs the pre-send authentication and warm-up checks described above [[4]](#sources). Sending, follow-ups and the mailboxes your plan includes use no credits [[4]](#sources).

If you want to see how the rest of the setup fits together, How Outlevo works explains the approval and kill-switch model.

Sources

  1. Google: Email sender guidelines
  2. Microsoft: Outlook requirements for high-volume senders
  3. Yahoo: Sender best practices
  4. Smartlead integration
  5. Smartlead, average cold email reply rate (2026)
  6. Smartlead, The State of Cold Email 2026: reply rate per contact (top quartile; January to June 2026, more than 850 million emails)
  7. RFC 7208: Sender Policy Framework (SPF)
  8. Managed mailboxes integration
  9. How Outlevo works

Related

See what Outlevo finds on your own website

Free, no sign-up, about a minute.