Cold Email Deliverability: A Spam Placement Diagnostic

Cold email deliverability fails for one of four reasons, in diagnostic order: broken authentication, damaged reputation, problematic sending patterns, or content that triggers filters. Authentication failures confirm in message headers within five minutes; reputation recovery takes weeks to months. Most diagnostic guides start with content and skip the first three causes. This article works through each layer in that sequence, separating what you can measure directly from what you can only infer. Google and Microsoft each publish free tools that surface the same signals those providers use to make filtering decisions.

What Cold Email Deliverability Actually Measures

Delivery, inbox placement, and open rate are three distinct measurements that most outbound teams treat as interchangeable. Delivery means the receiving mail server accepted your message via SMTP. Inbox placement means the message landed in the primary inbox rather than the spam folder. An open rate tells you nothing reliable about either.

You can measure delivery: your sending platform logs an acceptance receipt or a bounce code. You can test inbox placement by sending to seed mailboxes you control and checking the folder. What you cannot see directly is which filter rule acted on a message or why. Filter decision logs aren’t surfaced to senders.

A message can be delivered and still be placed in spam. SMTP acceptance is a technical handshake; folder placement is a scoring decision the provider makes after that handshake completes. The Authentication-Results header in any delivered message is the closest measurable window into that scoring decision.

The confusion between “my emails are sending” and “my emails are reaching inboxes” is the most expensive diagnostic mistake in cold outreach. A high delivery rate and a near-zero reply rate can coexist when messages are accepted but placed in spam. Start every diagnosis by sending a test message to a real inbox you control and reading the full headers.

Inbox placement also varies by recipient domain. A campaign hitting Gmail inboxes may be in the Outlook spam folder at the same time. Provider-specific tools exist for each major platform, and those signals don’t carry across providers.

Authentication Checks: SPF, DKIM, DMARC, and Alignment

Authentication is the foundation of cold email deliverability, not the ceiling. A clean authentication record tells receiving servers that your domain authorized this sending source. It doesn’t ensure inbox placement, but a failed record is an immediate, confirmed disqualifier that appears in the message headers.

SPF, defined in RFC 7208, is a DNS TXT record that lists which mail servers are authorized to send for your domain. When the receiving server checks the envelope sender, an SPF PASS means the source is on the authorized list. An SPF FAIL or SOFTFAIL appears immediately in the Authentication-Results header. A minimal SPF record looks like this:

v=spf1 include:_spf.yoursendingservice.com ~all

DKIM, defined in RFC 6376, adds a cryptographic signature to the message. The receiving server fetches your public key from DNS and verifies the signature. A DKIM PASS confirms the message was not altered in transit and that the signing domain authorized the send. A DKIM FAIL is equally visible in the Authentication-Results header.

DMARC adds alignment and policy enforcement to both checks. Per the DMARC specification, a message can carry an SPF PASS from a sending service and a DKIM PASS from a signed subdomain, yet still fail DMARC if the header From domain doesn’t align with either. This is the most common authentication trap in cold outreach: individual protocol checks pass, but alignment fails. A minimal DMARC record:

v=DMARC1; p=quarantine; rua=mailto:[email protected]

To check your current state, send a test message to a mailbox you control and read the Authentication-Results header. Microsoft’s header authentication documentation explains each field for Outlook-received mail. The whole check takes five minutes and rules out the fastest class of confirmed problems.

Reputation Signals: Domain, IP, and Mailbox Level

Reputation operates at three levels, and not all three are directly measurable. Start with what is.

Google Postmaster Tools measures domain reputation and IP reputation for messages delivered to Gmail accounts. It reports reputation on a four-point scale (bad, low, medium, high) and shows your spam rate as a percentage of messages Gmail users reported as spam. The tool is free. If your cold email deliverability issue is concentrated in Gmail inboxes, Postmaster Tools is the first measurement source to open. The limitation: your domain needs sufficient Gmail traffic before data populates.

Microsoft SNDS reports complaint rates and sender reputation for traffic to Outlook.com and Hotmail accounts. It’s also free and requires submitting your sending IP addresses to access. If your filtering issue is concentrated in Microsoft-hosted mailboxes, SNDS surfaces the complaint signals driving the problem.

What neither tool measures: reputation at Yahoo, Apple Mail, or any other provider. For those, you infer from bounce codes, postmaster feedback loops where available, and complaint data your sending platform returns. That absence of direct data doesn’t mean reputation isn’t a factor; it just means you’re working from indirect evidence.

Shared IP addresses add a variable you can’t control. If your ESP sends from an IP pool shared with other customers, poor behavior from those senders hits your reputation scores directly. A dedicated IP isolates your sending history but starts with no reputation record and needs a gradual warmup before high-volume campaigns can run.

Sending Patterns, Volume, and List Hygiene

The signals in this section are measurable. The thresholds at which they trigger provider filtering are not published by any major provider.

A sudden increase in sending volume from a domain or IP with no prior high-volume history is a pattern providers use to flag potential abuse. The result shows up as increased bounce rates and, with a delay, as declining reputation scores in Postmaster Tools and SNDS. Gradual volume increases give providers time to build a reputation model before a campaign scales.

Google’s bulk sender requirements include a mandatory one-click unsubscribe header for high-volume senders to Gmail. Missing that header on qualifying campaigns is a compliance gap that drives complaint rates up. Google’s published guidance flags sustained spam rates above 0.10 percent as a risk threshold and above 0.30 percent as requiring immediate correction.

These signals all feed into deliverability outcomes, but their relative weight depends on the provider and your current reputation baseline. Hard bounce rates consistently above 2 percent point to a significant chunk of the list containing invalid addresses. Complaint rates are just as concrete: every spam report from a Gmail or Outlook user feeds directly into the domain and IP reputation scores that Postmaster Tools and SNDS report back to senders.

Content Checks: Copy, Links, HTML, and Tracking

Content is the most discussed and least measurable variable in cold email deliverability. Address it after authentication, reputation, and sending patterns are confirmed clean.

URL reputation is one checkable signal. Links in your message are evaluated against blocklists maintained by providers and third parties. If a domain in your links appears on a known blocklist, that’s a measurable finding, even though the filter logic that uses it isn’t public. You can check any link domain against publicly maintained blocklist databases.

HTML structure is a documented filtering signal. Heavily templated messages with many images, external fonts, multiple tracked links, and complex nesting look to a filter like broadcast email rather than 1-to-1 correspondence. Plain-text messages or lightly formatted HTML with a single tracked link reduce the structural markers of a mass send.

Tracking pixel domains carry their own reputation. A pixel served from a shared tracking subdomain used by thousands of senders inherits the complaint history of all of them. If your ESP uses a default shared tracking domain, check its reputation in Postmaster Tools before blaming your copy or subject line.

Note: A seed mailbox test returns actual folder placement for a specific message sent to controlled addresses. A spam score tool returns a heuristic estimate based on message structure. These are different types of evidence. A passing spam score does not confirm inbox placement.

The “spammy words” category in most deliverability guides treats content filtering as a measurable threshold. It is not. No provider publishes its word list, and filter logic changes continuously. The only way to test copy’s effect on placement is a controlled seed mailbox comparison with a single variable changed between sends.

The Cold Email Deliverability Diagnostic Checklist

Run these steps in order. Every step before the last produces a confirmed finding. The final step produces an inference drawn from a controlled test.

Step 1: Read the Authentication-Results header. Send a test message to a mailbox you control at Gmail or Outlook. Open the raw headers and find the Authentication-Results field. SPF, DKIM, and DMARC each report pass or fail. A fail on any of them is your first confirmed problem and the fastest to fix.

Step 2: Open Google Postmaster Tools. Check domain reputation and the spam rate graph for Gmail traffic. A bad or low domain reputation rating means Gmail has flagged your sending domain. The tool is free; data takes 24 to 48 hours to populate after sufficient sending volume.

Step 3: Open Microsoft SNDS. Check the complaint rate and reputation rating for your sending IP on Outlook and Hotmail traffic. A red or yellow status means elevated complaint rates. Free to access after submitting your sending IP addresses.

Step 4: Run a seed mailbox test. Send a representative message to addresses you control across Gmail, Outlook, Yahoo, and any other providers your list includes. Check actual folder placement in each client. This is the only direct measurement of inbox placement across providers.

Step 5: Audit bounce and complaint data in your sending platform. Hard bounce rates above 2 percent and complaint rates trending upward are confirmed list quality problems with a direct link to the reputation signals both tools report.

Step 6: Examine content last. Run controlled seed tests comparing copy, link, and HTML variations with a single variable changed between sends. Any finding here is an inference drawn from a controlled test, not a measurement returned by a provider.

Cold email deliverability improves fastest when you start from confirmed evidence and move toward inference. Authentication failures fix in DNS within minutes. Reputation recovery takes weeks to months. Content iteration runs faster but without direct feedback from provider filter logic.

The most common diagnostic mistake is treating a low spam score as a clean bill of health while skipping the reputation check entirely. Reputation data is free, sourced directly from the providers making the filtering decisions, and more authoritative than any third-party heuristic estimate. Pull it before you rewrite the subject line.

Every threshold and requirement cited in this article is sourced to primary documentation: RFCs, Google’s published sender guidelines, or Microsoft’s own documentation. If a vendor or guide states a deliverability limit not found in one of those sources, treat it as an opinion, not a fact.

Prices, limits and product capabilities were checked on 2026-08-12 and change without notice. Nothing here is a prediction of results for your list, domain or market.