When you click send on an email, your mail client shows a quick spinner and then moves the message to the Sent folder. That animation is the entire extent of its involvement. Everything that happens next, the actual delivery of your message to another person's inbox, is infrastructure you never see and rarely think about. Until something goes wrong.
I work with email deliverability regularly, and the number of times I have heard "but I sent it" from someone whose message bounced, was rejected silently, or landed in spam is high enough that I wanted to write this down properly. Here is exactly what happens between the moment you click send and the moment someone reads your message.
Your mail client's job ends at the handoff
The first thing to understand is that your mail client, whether that is Gmail, Outlook, Apple Mail, or anything else, is not a postal service. It is a composition and display tool. When you click send, it takes your message and passes it to your outgoing mail server using a protocol called SMTP (Simple Mail Transfer Protocol). That handoff is your client's entire job.
Think of it like dropping a sealed envelope into a post box at the end of your street. You have done your part. What happens next is entirely in the hands of a postal network that operates independently of you.
Your SMTP server, usually operated by your email provider (Google, Microsoft, Fastmail, or whoever hosts your domain), accepts the message and queues it for delivery. This is where the real work starts.
MX records: how the internet finds the destination mailbox
Your mail server now needs to figure out where to send the message. It knows the recipient's address (say, someone@company.com), but it does not know which server handles email for that domain. That information lives in DNS.
DNS, the Domain Name System, is a massive distributed directory the internet uses to look things up. Among its many record types, the MX record (Mail Exchange) is the one that answers the question: "Which server should receive email for this domain?" A domain can have multiple MX records with priority values, so if the first server is unavailable, your server tries the next one.
Your outgoing server queries the DNS for the recipient's domain, retrieves the MX records, and now knows where to knock. If a domain has no MX records, delivery fails immediately. If a domain's MX records point to a server that is offline, delivery fails after retries. Most of this happens in under a second when everything is healthy.
The SMTP conversation between servers
Once your server has the destination's address, it opens a TCP connection to that server on port 25 (the standard SMTP port for server-to-server mail) and they have a brief, structured conversation. It goes roughly like this:
Your server identifies itself. The destination server confirms it is listening. Your server says it has a message from your address, destined for the recipient. The destination server either accepts or rejects it at this point. If it accepts, your server sends the message headers and body. The destination server acknowledges receipt. Connection closed.
That entire exchange is what "delivery" technically means at the protocol level. Your message is now on the destination server. But it is not in anyone's inbox yet. That decision still has to be made.
Authentication: SPF, DKIM, and DMARC
Before the destination server accepts your message into an inbox, it runs a set of authentication checks. These are designed to answer one question: is this message actually from who it claims to be from?
SPF (Sender Policy Framework) checks whether your sending server is authorized to send email for your domain. The domain owner publishes a DNS record listing the IP addresses or services allowed to send on their behalf. If your message arrives from a server not on that list, SPF fails.
DKIM (DomainKeys Identified Mail) is a cryptographic signature attached to every outgoing message. The sending server signs the message with a private key. The destination server retrieves the matching public key from DNS and verifies the signature. This confirms the message was not altered in transit and genuinely originated from an authorized sender. A missing or invalid signature is a DKIM failure.
DMARC (Domain-based Message Authentication, Reporting, and Conformance) is the policy record that tells destination servers what to do when SPF or DKIM fails: deliver anyway, quarantine to spam, or reject outright. It also includes alignment rules, checking that the domain in the From header matches the domain that passed SPF or DKIM.
All three of these records live in DNS. They are set up by the domain owner, not by any individual sender. If your domain's DNS is misconfigured, every message you send carries broken credentials, even if your message itself is completely legitimate. I have covered the setup process in detail in a separate piece on configuring SPF, DKIM, and DMARC.
Spam scoring: the last gate
Passing authentication does not guarantee inbox delivery. It just gets you through the security desk. The spam filter is a separate, often more complex layer.
Spam filters score incoming messages on dozens of signals: the reputation of the sending IP address, the reputation of the sending domain, the content of the message (certain words and patterns score higher), the ratio of text to links, whether the links point to known bad domains, the engagement history between sender and recipient, and complaint rates from previous recipients.
A message can pass SPF, DKIM, and DMARC perfectly and still land in spam if the sending domain has a poor reputation, or if the content triggers filters, or if past recipients marked similar messages as spam. Reputation is accumulated over time and lost faster than it is built. A newly registered domain sending a high volume of messages is a classic spam signal, regardless of content.
This is why warming a new sending domain matters: you start with low volume, build a track record of real engagement, and let reputation accumulate before scaling up. Skipping this step and blasting thousands of messages from a cold domain typically results in bulk foldering and deliverability problems that can take weeks to recover from.
What you can actually control
Most of what happens after you click send is determined long before that click. The decisions that matter are:
- SPF, DKIM, and DMARC configured correctly for every domain you send from. This is the baseline. Without it, major providers like Google and Microsoft will increasingly reject or junk your messages.
- Sending infrastructure with good IP reputation. Reputable providers (Google Workspace, Microsoft 365, Fastmail, or dedicated sending platforms like Instantly.ai or Smartlead) start with established IP reputations. Sending through a compromised or unknown IP is a quick path to deliverability problems.
- List hygiene. Sending to invalid addresses generates bounces. High bounce rates damage your sender reputation with every major mail provider. Remove hard bounces immediately and suppress unengaged addresses regularly.
- Complaint rates. Every recipient who marks your message as spam is a signal to their provider. Keep your complaint rate below roughly 0.1 percent to stay in good standing with Google and Microsoft. Above 0.3 percent and you will start seeing systematic filtering.
- Monitoring your reputation. Google Postmaster Tools and MXToolbox show how Google sees your domain and your sending IP. These are free, and most people do not check them until something breaks.
The frustrating truth is that most email problems are invisible. The sender sees "Sent." The recipient never sees the message. No error, no bounce, just silence. Understanding the infrastructure between those two points is what turns a guessing game into a diagnosis.
Questions about email deliverability or DNS setup? Get in touch.
Comments