How Email Works: Part Two

In part one we looked at what happens to an email marketing message from the point at which you hit “send” until it reaches the receiving site’s servers. We followed the message through the submission to the ESP’s MTA and then over SMTP to the ISP’s MTA. There are, however, several more steps to take before it reaches the end recipient.

The figure below shows the steps on the ISP side. The specific details will vary from ISP to ISP and this is a simplified explanation of the process. The reality for Gmail or involves thousands of servers spread across continents with failover, backup and adminstrative proceses. This, though, is conceptually how it

The first step is the spam analysis. In modern systems this process starts during the SMTP transaction and is used to inform the accept/refuse decision as well as the inbox/spam folder decision, i.e. whether the message will be accepted or bounce with a “550 Access Denied” or similar and whether it will make it to the inbox. We will come back to the spam analysis process, but for now let’s follow the message through to the recipient.

Once accepted the message is placed in a database and, if all went well, it wasn’t marked as spam and is now part of the recipient’s inbox. The next time the recipient checks their email it will be delivered to them. How this happens depends on whether they’re using a webmail interface such as the native Gmail,, Yahoo or AOL systems or whether they’re using an email application such as Outlook, or a mobile app.

In the case of the former, the email is “delivered” as part of a web page using the standard HyperText Transfer Protocol (HTTP). In the latter, the email client, technically known as a Mail User Agent (MUA), requests the message (and any other new messages) from the ISP’s Mail Delivery Agent (MDA) through an email-specific protocol. This is usually Post Office Protocol (POP) or the Internet Message Access Protocol (IMAP) that are sometimes referred to by including their version numbers as POP3 or IMAP4. POP and IMAP only handle receiving email. To send email a client typically uses SMTP, just like a mail server. I recognize these are somewhat esoteric terms that don’t typically come up in marketing conversations. If, though, you have reason to spend time talking with an email developer, deliverability consultant, postmaster or blocklist operator you may be very glad you know what they mean!

It’s worth mentioning that in many cases the email client versus web interface choice is not exclusive. The same individual may read a given message through both mechanisms depending on their device or location. How the email is viewed has a significant effect on rendering and may also affect results such as open rate, as we’ll see later.

At this point our message has been delivered but we’ve skipped or glossed over some pieces of the process that are extremely important to most email marketers. Specifically the burning question of “Will it get to the Inbox?” and also, “How do we know if it was opened or if links were clicked?”.

Next time we’ll explore both of those questions. We will also look more closely at the email message itself – the piece that in part one was just shown as {email message goes here}. You could be forgiven for thinking that is just the text (or HTML) of the message but there’s rather more to it than that. A significant quantity of information is included in the message that is not immediately visible in most email clients. Understanding this information is essential to understanding both deliverability and tracking. 

Related reading

Flat business devices communication with cloud services isolated on the light blue background.
Image from Desigual welcome email to new subscribers.