Guide to deliverability. Basics of email (SPAM) filters

If you are constantly sending emails, there is a slight chance that you will run into issues with spam filters at some point, and that’s okay.

According to a recent study by Returnpath, called 2018 Deliverability Benchmark, 14 percent of emails sent went straight into the SPAM section of the inbox.

Look, even if you’re not a spammer by definition, inbox providers are working very hard on allowing only relevant emails to go through to your inbox.  After all, if a particular inbox sorts your emails inconveniently, you’re more than likely to switch to their competitor that will do the job correctly.

In rare scenarios, spam filters can be triggered even if you’re emailing people that subscribed to your emails.

As you already could’ve guessed – this is what this post will be about – email filters.

We should note that this post was inspired by the Mailerlite Spamhaus incident. In essence, it was a failure to uphold their IP reputation by letting illicit senders use their system. That meant that even legit email marketers were sometimes mistakenly identified as spammers and their emails went to the spam section of the inbox. And since Mailerlite has a lot of customers, that hurt a lot of businesses.

But more on that later. Let’s get acquainted with email filters.

Email filters

Originally, email filters were designed to keep the spam out of your inbox, or place it in the spam section, at least. Today, most inbox providers use filters to organize the content you are receiving.

The filters discussed in this post are known as spam filters.

A spam filter essentially is a combination of multiple rules that look for telltale signs to help determine if the email is legit. There are multiple indicators and nobody, except for their creators knows what exactly they are looking into when determining the spam score.

If details were leaked, it would pretty much render any spam filtering useless and we would all get swarmed with spam. To illustrate the point, here’s a graph that shows the global spam volume as a percentage of total e-mail volume:

capture

Source: Statista

Now it’s not as bad as it used to be at its peak – the world had more than 7 spam emails for every 3 legit ones. If spam filters wouldn’t exist, every second email you receive would be a spam one. That’s why they’re pretty handy to have.  So how do they work?

Each indicator has a value assigned to it. The sum of all of the indicators results in an overall spam score. A high spam score means that the email will be nicely escorted into the spam section of the inbox.

Keep in mind that various inboxes have vastly different filters, which means that you may land in the spam section of a provider X, but land in the primary tab of a provider Y.

That is especially apparent when looking at the results of B2B email marketing – some business inboxes have strict filtering procedures, only allowing specific communications, blocking any outside emails, resulting in false-positive bounces or placements in the spam section.

New indicators are constantly being added and updated. It also takes into account what people report as spam or phishing in order to recognize future emails. Who knows, they may be implementing AI(Artificial Intelligence) or AGI(Artificial General Intelligence) to aid in spam recognition.

So let’s move on to what you should do to avoid triggering the spam filters.

Tips to avoid the spam section of the inbox

  • Spam filters want to know if you’re actually authorized to send emails on behalf of a domain, so you should only send through verified domains. To verify a domain at Sender.net, simply contact the live support for the SPF/DKIM records and they will be provided for free.
  • Some spam filters will be triggered by a poor IP reputation. If you don’t have a large subscriber list (over 50,000 email addresses), chances are you’re sending from a shared IP range (IP pool). That means there is a chance of you getting IP’s with a bad reputation if you’re not using a trustworthy email marketing provider. That’s why we at Sender.net work so hard on maintaining our IP reputation high and only approve accounts that are trustworthy. And if you want a dedicated IP so you could have total control of your reputation, we can provide it for you.

Here’s a recent incident to illustrate the point.

Mailerlite was blacklisted by Spamhaus.org on the 1st of August, 2018. Spamhaus claims that Mailerlite allowed Snowshoe Spamming through their domains.
Snowshoe spamming essentially means that spammers was spread out over several domains and IP addresses, meaning it was a lot more difficult to identify and stop. 

The strategy of snowshoe spamming is similar to actual snowshoes that distribute the weight of an individual over a wide area to avoid sinking into the snow. Likewise, snowshoe spamming delivers its weight over a wide area to remain clear of filters. – Source – Techopedia.com

Mailerlite then said that they’re addressing the issue and that:

‘Most of our current senders that were not abiding by the highest email marketing practices were put on hold.’

Which basically means that illicit senders are and were identifiable to them. The question is – if they’re easily identifiable, why were they not suspended in the first place? Was it greediness or negligence that lead to this downfall? Either way, it is completely possible that this was a ticking bomb they knew about.

So what’s the big deal, from the perspective of marketers?

Since most senders use shared IP’s, meaning people in the same IP pool have similar IP reputation, all emails sent through a blacklisted IP goes straight into the SPAM section, doesn’t matter if they’re the same person or not.

Ant that’s what happened in this case – Mailerlite’s IP addresses were blacklisted. And stayed like that for a long time(almost three weeks). That may not seem much, but for businesses that rely on email marketing for their sales, it’s pretty traumatic.

If you’re still worried about your IP reputation and just can’t get a good night’s sleep cause you’re worried about the IP reputation, there’s this thing called a dedicated IP.  You always get a dedicated IP if you have 50,000 subscribers or more. If you have less, you can pay extra for that. Contact our live customer support for more details.

  • Bad code can seem suspicious and cause issues with spam filters. Please use expert-crafted templates only, or choose from our template library for error-free templates.

Don’t link to sites that have established a bad reputation. Always investigate who you’re linking to if you’re linking to a website you don’t own or trust. Our tip is to use Mozbar from Moz. Make sure you check the Spamscore of the website.

If the spam score reaches more than 4-5, it’s definitely a suspicious site. Generally, the higher the spam score, the higher the chance of it being a spammy website – hence the name Spamscore.

  • It’s probably unnecessary to say this in the post-GDPR era of email marketing, but make sure that your subscribers have opted in to receive your emails. Cold-emailing will definitely result in poor engagement and in turn, in poor reputation.
  • Even if you’re not cold-emailing people, you still have to work for engagement. Design your emails in a way that encourages action and excitement. The design has to be aesthetic, the copy must be easy to read and the email has to have a clear call-to-action. Well, this rule doesn’t apply if it’s a transactional email. A transactional email can simply tell the details of something without using too much fancy design.

If your emails result in poor engagement, it demonstrates that the people you’re emailing aren’t willing to interact with you, which will result in a higher risk of landing into the SPAM section of the inbox, resulting in even worse engagement rates.

One more thing…

On rare occasions, the path to the inbox can be interrupted by things called Gateways. Imagine them as the goalkeepers of our inboxes. It can easily block the email from going out, which means it will never even reach the inbox or the spam folder.

So an email’s path to the inbox looks like this:

pasted-image-0Source

So, it basically works like this (in simple terms) – a malicious email will  be blocked at the gateway, which means that it never even reaches the inbox. If it’s malicious and still reaches the inbox, the spam filter will identify it and place it in straight to the spam folder.

Only after overcoming these obstacles, the email can reach the inbox.

Keep in mind that if you’re emailing lots of people that use Outlook, you’re approximately three times more likely to be placed in the spam folder.

Take a look at the global inboxing rate of the four most popular inbox providers:

Source

It seems that Outlook is a real pain in the butt for most marketers. Rightly so – everybody loves deliverability.

So the key thing is this:

If you’re doing legit email marketing and if you’re using a trustworthy email marketing provider, you’ve got only several things to worry about – mostly design issues. Spam filters should be the least of your concerns.

Unless your subscribers use Outlook. Then you’re screwed.

Onward & Upward,

Sender.net Team