Email Deliverability Fundamentals

Every discipline covered elsewhere in this playbook - segmentation, automation, cold outreach, measurement - assumes the email actually arrives somewhere a person will see it. Deliverability and sender reputation are what determine whether that assumption holds, and they're also the layer most commonly treated as an afterthought: copy gets polished, subject lines get tested, and none of it matters if the message never clears the inbox in the first place.

Domain Authentication Is the Non-Negotiable Starting Point

SPF, DKIM, and DMARC are the three authentication records that tell mailbox providers a sending domain is legitimately who it claims to be, rather than something spoofing a trusted name. SPF authorises which servers can send on a domain's behalf, DKIM cryptographically signs messages to prove they weren't altered in transit, and DMARC tells receiving servers what to do with messages that fail either check. Missing or misconfigured authentication doesn't just risk the occasional spam folder placement - major mailbox providers now require all three to be properly configured, or bulk senders risk being blocked outright regardless of how engaged their list is. This is genuinely the first thing to verify, before touching content or segmentation, because nothing else in this playbook compensates for authentication that isn't set up correctly.

Sender Reputation Is Earned Gradually and Lost the Same Way

Sender reputation is the trust score mailbox providers assign to a sending domain and IP, built from engagement signals over time - open and click rates, complaint rates, bounce rates, and how consistently a sending pattern behaves. It isn't a single number checked once; it's continuously reassessed, which is why deliverability problems rarely feel sudden even though they often get noticed suddenly. A domain sending consistently to an engaged, permission-based list (covered in the previous lesson) builds reputation gradually. A domain sending inconsistently, to a list full of disengaged or invalid addresses, erodes it just as gradually - until it crosses a threshold and inbox placement visibly drops.

New Domains Need to Be Warmed, Not Launched at Full Volume

A brand-new sending domain has no reputation yet, good or bad, and mailbox providers treat unfamiliar senders with more suspicion than established ones. Warming a domain means starting with small send volumes to a domain's most engaged contacts and gradually increasing over two to four weeks, letting positive engagement signals build reputation before volume ramps up. Sending high volume immediately on a new domain is one of the most common and most avoidable causes of early deliverability problems - it looks identical to spam behaviour from a mailbox provider's perspective, regardless of how legitimate the actual content is.

Bounces and Complaints Compound Faster Than They Look

Not all bounces are equal - a hard bounce versus a soft bounce require different handling. A hard bounce means the address is permanently invalid and should be removed from the list immediately; continuing to send to it repeatedly signals poor list hygiene to mailbox providers. A soft bounce is typically temporary - a full inbox, a server issue - and can be retried, but a soft bounce that persists across several sends should eventually be treated like a hard one. Spam complaints matter even more: a complaint rate above roughly 0.1% is enough to start damaging reputation at most major mailbox providers, which is a small enough number that a single insufficiently-targeted send to a disengaged segment can meaningfully hurt a domain's standing.

Diagnosing an Existing Deliverability Problem

When emails start landing in spam without an obvious cause, the diagnostic order that works is authentication first, then recent list hygiene, then recent engagement trends - not guessing at content or subject lines before ruling those out. For a deeper walkthrough of this specific scenario, see why your emails are landing in spam even though nothing changed: the most common answer is that something did change gradually - a rising complaint rate, a stale segment left in a send, an authentication record that lapsed after a domain or DNS change - and it simply crossed a threshold that made the effect visible all at once.

What's Next in This Playbook

With a permission-based list and solid deliverability foundation in place, the next lesson moves to lifecycle automation - how triggered flows like welcome sequences, cart abandonment, and win-back campaigns actually get built, and why they consistently outperform one-off broadcasts sent manually.