There’s no definitive answer on exactly what happens, for reasons I’ll explain in a moment.
However, some general concepts apply when you mark something as spam.
The first thing we need to know is whether you’re marking it as spam in an email program running on your machine, or if you’re using an email service’s web-based interface via your browser.
Marking spam in your email program
An email program is software that runs on your computer and downloads messages from your email service to your local hard disk. Examples of these kinds of email programs include Microsoft Office’s Outlook, Thunderbird, Windows Live Mail, and the Mail program included with Windows itself. There are dozens, if not hundreds, of others.
When you mark something as spam (or “junk”, as it’s called in some programs), you are typically telling only that program that the email message is unwanted. Particularly if your email account is downloaded using the POP3 protocol, the information about what you’ve marked as spam typically does not make its way back to your email service provider. The result is that it does not affect what email will continue to be downloaded in the future.
Dedicated email programs and IMAP
Mobile devices have added a couple of twists on whether or not marking something as spam in a computer-based email program will actually make it back to the email service.
Some email services now provide dedicated email programs. On my mobile phone, for example, is an app I use to access Gmail. Technically, that’s an email program (app) running on my computer (the phone). However, since it’s dedicated to handling Google Mail, when I mark something as spam using the program, the information is transmitted back to Google’s servers. Similarly, I have the Outlook mobile app, and when I use it to mark something as junk in my Hotmail account, that information also makes its way back to Microsoft’s servers.
IMAP is used in desktop email programs to allow you to access your email from multiple different devices and keep everything in sync. IMAP does this by leaving the “master copy” of your email on the mail server, and simply maintaining a synchronized copy of your mail on your computer. Change, delete, or move mail around in folders on your PC, and those changes will also be reflected in the master copy on the mail server.
When you mark an email as spam, some email programs simply move it into a spam or junk folder. If you’re using IMAP, when the move is reflected on the server’s master copy, it may be enough to notify the service that this message is spam.
There are no blanket rules, and – aside from dedicated apps, like the Gmail and Outlook apps – it’s difficult to make assumptions about exactly how your email program works with your email service when it comes to spam. The best I can suggest is to check the help information available for each.
How email programs use the information
An email program may use the fact that you’ve marked something as spam in several different ways:
- It might add the sender’s email address to a block list. (This is a separate function in some email programs.) Unfortunately, block lists based on email addresses are typically not at all effective in the war against spam – spammers are constantly changing or faking their email address.
- It might add the IP address of the sender (or the sender’s email service) to a block list. Once again, IP-address-based blocking is also not effective against spam; spammers send from millions of different IP addresses.
- It might analyze the contents of the message and identify various characteristics of the message it then records as “looking like” spam to you. When more email arrives with enough similar characteristics, it might be automatically flagged as spam. This is the most common, and currently the most effective, email-program-based, spam-filtering technique.
As effective as it can be, the problem with looking for characteristics is that it’s difficult to predict what will or will not constitute spam. You might get an email that is clearly spam to you, mark it as spam, and then later get another nearly identical message that was still not filtered.
These types of “learning” or adaptive filters don’t necessarily act immediately, but build up statistical characteristics of what spam looks like to you depending on additional factors, like the number of times you mark similar messages as spam. It might not be until the second or third (or fourth or tenth) time you mark a particular type of spam that the filter will have enough confidence to kick in and automatically identify similar messages as spam in the future.
“Similar” is, after all, a fuzzy concept.
What email programs do with spam
Great, your email program has successfully identified something as spam – now what does it do with it?
As I discussed above, most email programs do nothing more than move the email to a Junk or Spam folder. That’s it. Period.
There’s no notification back to the sender, and no notification to the email service. Everything happened on your computer and only on your computer.
This type of spam filtering is nothing more than placing email detected as likely being spam into a different folder as it’s downloaded to your computer.
Marking spam on an email service
If you’re using a web browser (like Internet Explorer, Chrome, or others) to read your email, you’re using an online email service. Your email is stored “in the cloud,” not on your PC, and you’re simply viewing it via a web-based interface.
Examples of web-based email services include Outlook.com, Gmail, Yahoo Mail, and many, many others. Your own ISP or email service provider may also have a web-based interface for your email, in addition to the services that allow you to access it via your PC.
The important thing here is that you’re using your web browser to interact with your email stored on your email service’s server.
How email services use the information
When you mark something as spam on an online service, you’re doing essentially the same thing you did above with a PC-based email program – you’re telling the service, “I think email that looks like this is spam.”
The difference is that you and every other user of that service are all telling the provider what you think is and is not spam.
Exactly how the service provider uses that information is a mystery, and that’s on purpose.
For one thing, they don’t want spammers to learn the details of the mystery; that would make it easier for spammers to know how to work around it. For another, how service providers use that information is constantly changing in response to the ever-changing nature of spam.
There are several approaches email service providers may or may not use:
- Things that you mark as spam are used to identify and filter spam only for you. This is basically the PC model at the server level – you’re not impacted by the spam decisions of other users.
- Your marking something as spam goes into a single database used for everyone. Only things that everyone thinks look like spam are actually filtered. No matter how often you mark something as spam, if everyone else on the service treats it as legitimate, you may never see it filtered.
- Hybrid: a combination of what everyone thinks is spam plus what you think is spam is used when filtering email destined for your Inbox.
My sense is that it’s mostly the later, hybrid approach, but as I said, it varies from provider to provider and changes over time.
What email services do with spam
Much like the email program on your PC, when spam is filtered by an email service, it is typically sent into a junk mail folder in your account. You can usually safely ignore that folder, and/or periodically check it for false positives.
No notice is sent to the sender. The email has, in fact, been delivered; it’s just been delivered to your Spam folder.
Some services take things a step further.
Some services identify spam at a global level – perhaps based on content or source or something else – and block it from being delivered entirely. You’d never see it in your Inbox or in your Spam folder. From your perspective, it’s like the email was never sent.
Occasionally, in these cases, the service sends back a bounce message to the sender.
Maybe. Once again, there’s no guarantee.
Regardless of whether you use PC- or web-based email, spam detection is an inexact and ever-changing science. You’ll see email that you consider to be spam – perhaps even obviously spam to you – be delivered into your Inbox.
So, you mark it as spam and your email program or service learns a little more about what you consider to be spam. Occasionally, however, legitimate mail will get marked as spam and be filtered. This is referred to as a “false positive.”
For anything that’s been filtered and placed into a Spam folder, you should have the ability to say, “This is NOT spam”. This is perhaps even more important than identifying spam.
Once again, the program or service takes this indication from you and learns from you that email like this should not be considered spam.
That’s an important step to take. Every so often, spend a few minutes in your Spam folder looking for things that were filtered and should not have been. Mark those as “not spam” to reduce the chance of similar mail in the future also being filtered.