Email deliverability is whether your message reaches the inbox. It’s not whether the receiving server accepted it. A mail can be delivered and still land in spam.
Deliverability is the share that arrives where someone will actually see it.
3 things decide it: your sending reputation, your authentication, and the quality of your list.
That distinction is the whole subject. Almost every deliverability problem people bring us turns out to be a delivery success and a placement failure. The two have completely different fixes.
What Is Email Deliverability?
Email deliverability refers to the percentage of sent emails that successfully reach a recipient’s inbox rather than being sent to spam or blocked entirely. Deliverability is the outcome of a judgement made about you before your email is shown to anyone.
Gmail, Outlook and every other mailbox provider decide, the moment your mail arrives, whether the sender behind it has earned the inbox. They’re not reading your email and deciding it’s good. They’re reading your history and deciding you’re safe.
That judgement is built from things you control. Whether your domain is authenticated. Whether the addresses you send to are real. Whether people reply or complain. Whether your volume looks like a person or a machine.
Deliverability vs Delivery vs Inbox Placement
Three terms get used interchangeably and they mean different things.
Getting them apart is the first practical step, because each fails for a different reason and is fixed differently.
| Term | What it measures | What it tells you |
|---|---|---|
| Delivery | The receiving server accepted the message rather than bouncing it | Only that the address exists and the server did not reject you |
| Inbox placement | Where the accepted message was filed: primary, promotions, or spam | Whether a human will plausibly see it |
| Deliverability | Your ability to reach the inbox consistently, over time | Whether the problem is this campaign or your reputation |
A 99% delivery rate with 40% inbox placement is a normal set of numbers. It also describes a sender in trouble.
Most dashboards report the first figure prominently and the second not at all. That’s why people are often surprised to learn they have a problem.
Why Deliverability Breaks for Cold Senders and Not for Marketers
Most deliverability advice is written for email marketers.
It assumes something a cold sender doesn’t have: a list of people who asked to hear from you.
That one difference changes everything downstream. A marketer sends to subscribers who opted in. The domain has been sending for years, at a volume that barely moves week to week.
Their engagement signals are positive by default. When their deliverability slips, it’s usually a content or list-hygiene problem sitting on top of a healthy base.
A cold sender has none of that. The domain is often new. The recipients didn’t ask. Volume goes from zero to hundreds in a fortnight.
Every signal a provider uses to decide whether you’re trustworthy is either missing or looks, at first glance, like a spammer’s profile.
So the advice transfers badly. “Clean your list” assumes you’ve engagement data to clean it with. “Warm up gradually” means something different when you’ve no sending history at all. The mechanics are the same. The starting position isn’t.
What Actually Breaks Deliverability
Six things account for nearly every case we see. They’re listed in rough order of how often they’re the real cause rather than the suspected one.
Authentication That Is Missing or Misaligned
SPF, DKIM and DMARC are how a provider ties a message to a sender it can hold accountable. Without them, good behaviour can’t be credited to you and bad behaviour can’t be ruled out.
This is the cheapest problem to fix and the most common one to find.
Here is our guide: How to easily set up your SPF, DKIM, and DMARC records.
A Damaged or Absent Sending Reputation
Reputation attaches to your domain and your IP. It’s built from months of behaviour rather than days.
A new domain has no reputation, which isn’t the same as a good one. It means the provider has no reason to trust you yet.
Bounces
Sending to addresses that don’t exist signals that you didn’t check. Providers treat a high bounce rate as evidence of a list you didn’t build carefully.
It’s one of the fastest ways to lose standing. How to fix bounced emails, and prevent them covers the mechanics.
List Quality
Beyond bounces, the deeper risk is spam traps. These are addresses that exist only to catch senders who acquired a list rather than earned it.
A trap hit isn’t a small penalty, because it is close to proof of how the address was obtained.
Volume and Pattern
A jump from nothing to hundreds a day reads as automation. So does sending at exactly the same second every morning. Providers are pattern-matching against what human correspondence looks like.
Content, Though Less Than People Think
Spam-word lists are largely folklore at this point. Content is a weak signal next to reputation. It still matters at the margins.
We keep a list if you want it: 100+ spam words to avoid in your marketing and cold emails.
Shared IP Addresses
If you send through a shared IP, part of your deliverability is decided by strangers. Your mail inherits the reputation of everyone else on that address. That’s fine while they behave and expensive when they don’t.
The honest position is that this is a trade rather than a mistake.
A shared IP lends a small sender an established reputation, which is genuinely valuable when you’ve none.
A dedicated IP gives you control and a reputation nobody else can damage. But it starts empty, and it needs consistent volume to build one. Below roughly a few thousand mails a day, a dedicated IP usually makes deliverability worse.
How to Measure Email Deliverability
This is where most guidance is out of date, including advice we published ourselves. Two of the three tools everybody recommends have changed materially. One of them was never built for the kind of sending most of our readers do.
Google Postmaster Tools
Postmaster Tools still exists and is worth connecting.
It isn’t the tool most articles describe. Version 1 was retired on 30 September 2025, and the reputation dashboards went with it. Those were the domain and IP graphs every deliverability guide screenshots.
What remains is useful and narrower. Spam complaint rates, authentication pass rates, and delivery errors for mail you send to Gmail addresses. That’s real signal, and the complaint rate in particular is worth watching closely.
If an article tells you to check your Postmaster reputation graph, it hasn’t been updated in over a year.
One structural limit applies regardless. It only reports on mail sent to Gmail, so it tells you nothing about how Outlook treats you.
Microsoft SNDS, and Why It Cannot See B2B
Microsoft’s Smart Network Data Services is the standard recommendation for the Outlook side. For most B2B senders it’s close to useless, and almost nobody says so.
SNDS reports on IP addresses. It only measures mail delivered to individual Outlook and Hotmail consumer accounts. Mail sent to a business running Microsoft 365 doesn’t appear in it at all.
Since B2B cold outreach lands almost entirely in Microsoft 365 business tenants, SNDS is watching a population you aren’t sending to.
It’s also IP-level rather than domain-level.
If you send through a shared IP, the number you’re looking at isn’t really yours to interpret.
None of that makes it a bad tool. If you send consumer marketing to Outlook and Hotmail addresses from an IP you control, SNDS is genuinely valuable.
It’s a B2C instrument. The advice to “monitor SNDS” isn’t wrong, it’s addressed to somebody else.
Worth noting too that Microsoft moved the portal in June 2026.
The IP data now sits behind an authenticated API, so older instructions won’t work as written.
Seed-List Placement Testing
With Postmaster diminished and SNDS aimed elsewhere, seed-list testing is what’s left for measuring placement directly.
The method is simple. You send your real campaign to a set of addresses you control, spread across Gmail, Outlook, Yahoo and others.
Then you check which folder each one landed in. It gives you the number the other tools don’t: inbox versus spam, by provider, for this campaign.
Its limitation is equally clear. Seed addresses aren’t real recipients. They don’t open, reply or move mail out of spam, so their engagement profile is artificial.
A seed test shows how a provider treats your mail before human behaviour is factored in. It’s a good directional instrument and a poor absolute one. Anyone selling it as a definitive placement rate is overstating it.
Sender Score and MXToolbox
Sender Score grades a sending IP from 0 to 100 on observed behaviour. It’s a reasonable early warning for infrastructure problems. It’s also IP-level, so on shared infrastructure it describes your neighbours as much as you.
MXToolbox is the fastest way to check whether your domain or IP sits on a public blocklist.
It also confirms your DNS records resolve the way you think they do. We’ve written a detailed MXToolbox review if you want to know what it covers.
Neither tells you about inbox placement.
They tell you whether something is broken, which is a different and still useful question.
The honest summary is that no single tool measures deliverability. You triangulate. Complaint and authentication data from Postmaster. Blocklist and DNS checks from MXToolbox. Placement direction from seed testing.
And the part people skip: your own reply and bounce rates, the only signals generated by real recipients.
Our guide to email deliverability metrics goes through what to track and at what thresholds.
What Deliverability Looks Like for Cold Email
Two things are specific to cold sending and neither gets discussed much.
Placement differs sharply by provider. We measured this across 3 to 5 August 2026, on paid accounts. Mail in our warmup network reached the inbox 91.75% of the time at Gmail and Google Workspace addresses.
At Outlook and Microsoft 365 addresses it reached the inbox 77.97% of the time. That’s a gap of nearly fourteen points, between the two providers’ most B2B outreach targets.
If your prospect list skews toward Microsoft, your realistic ceiling is lower than the numbers in most deliverability articles. It’s not because you did anything wrong.
Measured on warmup traffic from paid TrulyInbox accounts, 3 to 5 August 2026. Warmup mail is sent between consenting mailboxes. It reads how providers treat a sender’s reputation, rather than predicting how your cold campaign will perform.
Reputation isn’t a state you reach. It decays. A domain that sat unused for two months is not where you left it.
The fix is the same as the start: consistent, moderate sending that gives providers current evidence you’re the sender they trusted before. This is what warmup is for, and why it runs alongside outreach rather than finishing before it.
How to Improve It
The short version, in the order that usually matters. Authenticate the domain properly. Verify the list before you send it to them. Keep daily volume steady rather than spiky.
Use a separate domain for outreach, so a problem can’t reach your billing mail.
And warm up your email account before and during the campaign, not only at the start. Each deserves more room than a paragraph.
Read our guide on: How to improve and maintain a high email deliverability rate.
FAQ
1. What Determines Email Deliverability?
Sender reputation first, then authentication, then list quality, then sending behaviour. Content matters least of the five despite getting the most attention.
All are evaluated together and weighted by history.
That’s why a new domain and an established one can send identical mail to different outcomes.
2. What Hurts Email Deliverability the Most?
Sending to addresses that don’t exist. A high bounce rate is both an immediate problem and a statement about how the list was built.
Providers respond to it faster than to almost anything else. Spam trap hits are worse, but rarer.
3. What Is an Email Deliverability Audit, and How Do I Run One?
It’s a structured check of the things that decide placement. Authentication records and their alignment. Blocklist status for your domain and IP. Bounce and complaint rates over recent sends. List verification.
Sending volume and consistency. And a placement test across the providers you actually send to. Run it before a campaign rather than during one, and again whenever you add a domain or change infrastructure.
3. Is Deliverability Different for Cold Email Than for Marketing Email?
The mechanics are identical. The starting position isn’t. Marketing email has opt-in recipients, engagement history and an established domain.
Cold email usually has none of the three, so the same rules produce harsher outcomes.
Advice written for marketers assumes a base of trust a cold sender has to build first.
4. How Long Does It Take to Fix Poor Deliverability?
It depends what broke. Authentication problems resolve within days of being fixed, because the signal is binary and current.
Reputation damage takes weeks, because reputation is built from sustained behaviour and providers revise it slowly by design. If someone promises to restore a burned domain in days, they’re describing an authentication fix.
5. What Does Google Postmaster Tools Actually Show Now?
Spam complaint rates, authentication results and delivery errors for mail sent to Gmail addresses.
It no longer shows the domain and IP reputation dashboards most guides screenshot; those went with the version 1 retirement on 30 September 2025. It remains worth connecting, and it remains blind to every provider that’s not Gmail.
If you would rather not manage the sending by hand, TrulyInbox is an automated email warm-up tool. It keeps volume consistent while you run outreach.
Deliverability isn’t a setting you configure once. It’s a running judgement mailbox providers make about you, rebuilt from evidence you supply every day you send.
The practical consequence is that almost everything worth doing is unglamorous and continuous. Authenticate properly. Send to addresses that exist. Keep volume steady. Watch the small number of signals that come from real recipients rather than dashboards.
Senders with good deliverability are rarely doing anything clever.
They’re doing the ordinary things consistently, and they started before they needed to.
