Email Deliverability
Email deliverability is the measure of how reliably your emails reach recipients' inboxes rather than landing in spam folders or being blocked entirely.
Key takeaways
- Delivery means accepted by the server; placement decides inbox, promotions or spam.
- Complaints damage reputation far more per event than opens improve it.
- Send marketing mail from a subdomain so transactional delivery stays protected.
- Placement varies by provider, so one blended figure conceals the real problem.
- Removing unengaged contacts raises placement more reliably than rewriting the content.
In depth
Deliverability is decided at the receiving end, by the mailbox provider, after the message has already been accepted. Delivery only means the receiving server did not reject it; placement decides whether it lands in the inbox, a promotions tab, or the spam folder. That decision draws on authentication results for SPF, DKIM and DMARC, the reputation attached to your domain and sending IP, the technical shape of the message, and above all how recipients on that provider have treated your mail recently.
Reputation moves on the ratio of positive to negative signals. Opens, replies, moving a message out of spam and adding the sender to contacts push it up; complaints, deletions without reading, and hard bounces push it down, and complaints weigh far more per event than opens do. Volume interacts with all of it: a sudden jump in daily sending from a domain with little history looks like a compromised account, which is why warming means raising volume in small steps.
Practical work splits into setup and hygiene. Setup is one-off: publish the authentication records, use a subdomain for marketing mail so a bad campaign cannot damage transactional delivery, and keep a working unsubscribe in every message. Hygiene is continuous: verify new addresses, remove hard bounces immediately, and suppress contacts who have not engaged for months. A scorecard quiz helps on the input side, because someone who fills in several questions to see their result gives a real address they will check.
Deliverability is provider-specific, so a single overall number hides what matters. Mail can sit in the inbox at one provider and in spam at another on the same day, and a corporate gateway may apply rules no consumer provider uses. Seed-list tests approximate placement but do not reproduce the engagement history of real recipients. Good deliverability also cannot make an unwanted message perform; reaching the inbox is a precondition for results, not a cause of them.
Example in practice
How to measure it
Start with the numbers your platform already has: delivery rate, hard and soft bounce rates, and complaint rate, the last expressed per thousand delivered because the acceptable level is very small. Read them per mailbox provider, not blended. A complaint rate that is fine overall but concentrated at one provider tells you where the problem is and which segment produced it.
Delivery rate alone will not show spam placement, so add inbox placement testing with seed accounts and, where available, provider postmaster dashboards that report your domain reputation directly. Then use behaviour as the honest proxy: a sustained fall in click rate with no change to the content usually means placement moved, not that the copy suddenly got worse.
Common mistakes
The most common technical error is publishing SPF and DKIM and stopping there, with DMARC left absent or on a permissive policy nobody reviews. Reports then go unread, an unauthorised sender using your domain goes unnoticed, and providers treat the domain more cautiously than they otherwise would. Set DMARC to monitoring first, read the reports until every legitimate sender is aligned, and only then tighten the policy.
The behavioural mistake is mailing everyone to protect the size of the list. Sending to contacts who have ignored a year of campaigns adds almost no revenue while producing the exact signals providers use to demote you, and it drags placement down for the people who do want the mail. Cutting an unengaged third of a list usually raises inbox placement and absolute clicks at the same time.
Frequently asked questions
What hurts email deliverability the most?
The biggest threats are sending to stale or purchased lists, high spam complaints, hard bounces, and missing authentication. Each one damages sender reputation and can trigger spam filtering or a blocklist entry.
How do SPF, DKIM, and DMARC help?
These authentication standards prove that your email genuinely comes from your domain and has not been spoofed. Setting all three up correctly is foundational, because mailbox providers distrust and often filter unauthenticated mail.
What is the difference between delivery rate and deliverability?
Delivery rate measures acceptance: the share of messages the receiving server took rather than bounced. Deliverability, or inbox placement, measures where accepted mail actually landed. A campaign can show a delivery rate near perfect while most of it sits in spam folders, which is why platforms reporting only delivery give a falsely reassuring picture of performance.
How do you warm up a new sending domain?
Start with a small daily volume to your most engaged contacts and increase it gradually over several weeks, roughly doubling every few days while the numbers stay clean. The point is to accumulate positive engagement before providers see large volume. Pause the increase whenever complaints or bounces rise, and never move an existing list onto a new domain in one send.
Why did my emails suddenly start going to spam?
Look first at what changed in the days before: a new list imported, a jump in volume, a change of sending platform, or an authentication record edited. Content changes matter far less than these. Check bounce and complaint rates by provider to locate the failure, then pause the segment responsible rather than continuing to send while you investigate.
Do spam trigger words still hurt deliverability?
Far less than reputation does. Modern filters weigh authentication, sending history and recipient behaviour much more heavily than individual words, so a trusted sender can use the word free without consequence. Content still matters where it correlates with bad senders: image-only messages, link shorteners, mismatched display names and hidden text all raise suspicion regardless of your reputation.
Should you use a dedicated IP address?
Only at consistent volume, roughly tens of thousands of messages a month or more. Below that, a dedicated IP never accumulates enough history for providers to judge it, and gaps between sends let the reputation reset. Smaller senders do better on a well-managed shared pool, where reputation is carried by the aggregate rather than by one irregular sender.
How does list building affect deliverability?
It sets the ceiling for everything that follows. Addresses collected with confirmed intent open and click, which builds reputation; addresses from purchases, contests or unverified forms produce bounces, complaints and spam-trap hits. A form that requires several deliberate steps, such as completing a quiz to receive a result, filters out typos and disposable addresses before they enter the list.