Google Postmaster Tools for Cold Email: Reading Metrics

Google Postmaster Tools is the only place Gmail tells you the truth about your sending. Every other deliverability signal you have access to is inference: seed tests, open rates, reply rates, third-party inbox placement scores. Postmaster Tools reports what Gmail itself recorded.

Open rates are the signal most often over-read, and why an open event stopped meaning much compares the six tracking tools that still report it.

It is also narrower than most people assume, and for cold email specifically it has a structural blind spot worth understanding before you rely on it.

Postmaster Tools v2 and the retiring reputation dashboards

Google launched a new version of Postmaster Tools, v2, in 2024 and is moving senders onto it. Its deprecation page says every dashboard in the old interface is available in v2 except Domain Reputation and IP Reputation, "which will be retired", and the v2 API leaves both out. Google has postponed switching off the legacy interface, so the reputation dashboards can still be read there for now, but they are on their way out. Google's stated reasons are that reputation data is not easily actionable, that changes in sender behaviour are slow to reflect in it, and that it can mislead because reputation is only one of many factors in deliverability.

For cold email that moves the weekly read. The Compliance status dashboard, v2's headline view, and the spam rate take the place the four reputation tiers used to hold, with delivery errors as the tie-breaker.

Legacy interface (v1)

• Still reachable: Google postponed its retirement
• The only place Domain and IP Reputation still show
• Its v1 API is being retired

Postmaster Tools v2

• Every other dashboard, plus Compliance status
• Report a delivery issue to Google
• A v2 API with compliance and domain endpoints

What it covers and what it does not

Postmaster Tools reports on messages sent to personal Gmail accounts, meaning @gmail.com and @googlemail.com addresses. Mail you send to Google Workspace business addresses does not appear. Most B2B cold email lands in Workspace inboxes, so the data you see is a sample of your consumer-domain traffic rather than a picture of your whole campaign.

That sample is still the best proxy available, because Gmail applies the same core reputation machinery to both. Treat it as an early warning system rather than a full report card, and pair it with bounce rate and reply rate data from your sending platform.

Setup takes minutes: add the domain you authenticate with, meaning the DKIM (d=) domain or the SPF Return-Path domain, and verify ownership with the TXT or CNAME record Google provides. Add every sending domain you own, not just the primary one, and add subdomains separately if you want their dashboards apart from the parent's.

The dashboards, one by one

Dashboard	What it shows	In v2
Compliance status	Each sender requirement: Compliant, Needs work or No data found	Yes
Spam rate	Share of inbox-delivered mail marked as spam	Yes
Domain reputation	A quality rating for your sending domain	No, retiring
IP reputation	A quality rating for your sending IPs	No, retiring
Authentication	Share passing SPF, DKIM and DMARC	Yes
Encryption	Share sent over TLS	Yes
Delivery errors	Share rejected or temporarily failed, by reason	Yes
Feedback loop	Tagged campaigns marked as spam	Yes

Compliance status

Each of Gmail's sender requirements gets one of three states: Compliant, Needs work, or No data found. This is the fastest triage screen in the product. If authentication, TLS, PTR records, unsubscribe headers, or spam rate are out of line, this dashboard names which one before you have to interpret any graphs.

"No data found" is common for cold email domains and usually means volume to personal Gmail is too low for Google to evaluate, not that something is broken. After you fix a Needs work item, Google says the change can take up to 7 days to show.

Spam rate

Google defines it as the percent of your messages that are delivered to an engaged recipient's inbox and then marked as spam by that recipient, and it counts only DKIM-authenticated mail.

Two implications people miss. First, the denominator is inbox-delivered mail to engaged recipients, so messages already filtered to spam do not count. A collapsing spam rate can mean your mail stopped reaching the inbox at all. Second, it is a recipient action, so it measures how your copy and targeting land, not your infrastructure.

The thresholds: Gmail's sender guidelines require every sender to keep the Postmaster Tools spam rate below 0.3 percent, and recommend staying below 0.10 percent and never reaching 0.30 percent. For cold outbound, treat 0.10 percent as the ceiling, not the target. Compare against cold email spam rate benchmarks before deciding whether a spike is yours or seasonal.

Domain and IP reputation, while they last

Both still show in the legacy interface and use the same four-tier scale. Google's own definitions, useful for as long as the dashboards remain:

Domain reputation follows the From domain. IP reputation follows the sending IP, which on shared infrastructure you do not control alone. If your domain reputation is High while IP reputation is Medium, the problem belongs to your provider's neighbourhood, not your list.

The tier is the wrong thing to watch day to day. The direction is the right thing. A domain sitting at Medium and climbing is healthier than one sitting at High and sliding.

Authentication

Percentage of your mail passing SPF, DKIM, and DMARC. Anything below 100 percent on a domain you fully control means a sending source is misconfigured. Find it before it costs you anything, using the SPF, DKIM, and DMARC setup guide and your DMARC aggregate reports.

Encryption

Percentage of authenticated traffic carried over TLS, inbound and outbound. Modern providers are at or near 100 percent. A dip points at a misconfigured relay.

Delivery errors

Percentage of authenticated mail rejected or temporarily failed, broken out by reason. This is where rate limiting, policy blocks, and bad recipients surface with Gmail's own wording. When a campaign goes quiet, this dashboard tells you whether Gmail is throttling you before your platform's bounce log does.

Feedback loop

Spam rates for campaigns tagged with a feedback loop identifier. Only useful if your sending platform sets FBL headers, which most cold email platforms do not. Skip it unless yours does.

1

Compliance status

Fix any requirement marked Needs work first. Changes can take up to 7 days to show.

2

Spam rate

Read it as a direction against 0.10 percent, and treat a falling rate beside falling volume as a warning.

3

Delivery errors

Rejections and temporary failures by reason: this is where throttling shows first.

4

Authentication

100 percent on domains you control; anything less is a misconfigured source.

5

Report a delivery issue

When all four are clean and mail still lands in spam, file it from v2.

Report a delivery issue

Postmaster Tools v2 lets a domain owner report mail that is wrongly classified as spam or phishing, rejected, or failing temporarily. Google only takes the report when the domain is verified, you are its owner, the domain meets the sender guidelines, and the message you report is DKIM and SPF authenticated with a From domain that matches. That makes it a last step rather than a first one: clean compliance and authentication are the entry ticket.

What to do at each reputation tier

Where the legacy dashboards still show a tier, the actions below apply; where they do not, run the same ladder on spam rate and delivery errors. The honest operating rule is that reputation is a lagging indicator with a slow recovery curve. By the time a domain reads Low, the sends that caused it happened weeks ago.

• High and stable. Keep going. Increase volume in small steps and watch spam rate rather than reputation for the first signal.
• High and falling. Look at what changed 7 to 14 days ago: a new list source, a new copy angle, a volume jump. Roll that change back before the tier moves.
• Medium. Halve volume on that domain, cut the weakest list segment, and verify everything you are about to send. Do not add new copy variables while you are diagnosing.
• Low. Stop cold sending from the domain entirely. Some operators try to warm their way out; in practice the calendar cost usually exceeds the cost of a replacement domain.
• Bad. Retire it and make sure the replacement domain does not share the same list, copy, or sending pattern that got you here.

This is why we maintain 3x sending capacity for every client campaign. Domains degrade, and the operators who survive it are the ones with spare capacity already warmed.

How often to look, and what to look at first

A dashboard nobody opens on a schedule is a dashboard that gets read after the damage. The cadence that works for cold outbound is weekly for the portfolio and daily for anything already moving.

Weekly, across every domain. Open compliance status first, because it names the failing requirement without asking you to interpret a graph. Then read spam rate as a direction rather than a level: three flat weeks followed by a reading half again as high is a trend worth acting on, even when every one of those readings sits inside Gmail's recommended ceiling.

Daily, on any domain that moved. A domain that dropped a reputation tier, or whose delivery errors spiked, gets read every morning until it settles. What you are watching for is whether the change you made had any effect, and reputation moves slowly enough that a week of daily reads is normal.

Read the graphs together rather than one at a time. The combinations carry more than the individual numbers. Authentication below 100 percent alongside a delivery error spike points at a misconfigured sending source. A falling spam rate alongside falling delivery volume is not an improvement: it usually means less of your mail is reaching an inbox where anyone could complain about it.

Write down what you changed and when. Reputation is a lagging indicator with a slow recovery curve, so the only way to attribute a change is to have a record of what happened one to two weeks before it. A one-line change log per domain is enough and it is the difference between diagnosing and guessing.

The cold email blind spot

Postmaster Tools was built for bulk senders running high volume through few domains. Cold email is the inverse: low volume across many domains. Two consequences.

First, per-domain volume to personal Gmail is often below Google's reporting threshold, so you get sparse data or none at all. Google withholds data on low-volume days to protect user privacy. The fix is to add every domain and read them as a portfolio: if three of your twelve domains show Medium, that is a signal even when the other nine show nothing.

Second, reputation is domain-level, so a single bad list dropped onto one domain does not poison the rest. That is the argument for domain separation, and it only works if you are actually monitoring each domain rather than assuming the pool is fine.

Running the portfolio rather than the domain

Reading twelve domains one at a time is how the pattern gets missed. Two habits make a portfolio readable.

Group domains by what they have in common: the list source they sent to, the copy angle they carried, the provider they were bought from, the week they were warmed. A problem that follows one of those groupings is a problem with that thing rather than with the domains, and it is invisible when each domain is read on its own.

Then keep spare warmed capacity, so that retiring a domain is a decision rather than an emergency. A pool with no slack forces you to keep sending from a domain you have already diagnosed as failing, which is how a Medium becomes a Low.

Pair the dashboards with a blacklist check whenever delivery errors spike, keep verification tight enough that bounces never contribute, and keep your compliance footing solid so complaints stay rare in the first place. For the wider picture, the deliverability guide covers what to change once the data tells you something is wrong.

Want someone watching these dashboards across your whole domain pool? Get a free campaign plan and we will show you how the monitoring is set up.