Most often we get customers who complain that their emails are going to spam. Even a simple Hi to a known contact ends up in spam.
They often ask us how do they not end up in spam and how doeshelp them. Before I answer that question there is something else I would want you to focus on.
If you are doing cold emailing or email marketing, it is extremely important that you know how spam filters work.
Before writing any further on how they function, a heads-up, no one person can tell you exactly how they function. Apart from learning from the blogs of email clients, analyzing data of millions of emails, we can tell with some degree of accuracy on how they function.
Spam filters of different email clients work on advanced AI and ML algorithms.
A few years back, I was talking to a couple of Microsoft Research team members who were working on Machine Learning and Natural Language Processing. They have algorithms to detect patterns, summarize content and can do sentiment analysis on whether the content in an email is spammy or promotional in nature. They work on huge databases which helps them to train their machines very fast and improve their efficiency in shorter time.
Apart from these, spam filters also check for a number of things like:
- IP Reputation
- Domain Reputation
- Sender’s Reputation
- Sending Infrastructure (SPF, DKIM, DMARC)
- Volume (frequency and sending limits)
- Spam Complaint Rate
- Hard and soft bounces
- Email content
- IP Reputation: Whenever you are sending an email, it will be routed through an mail server. And every mail server on the internet has an IP address associated with it. When you are sending an email, it contains your ip address of your mail server and goes to the client.Whenever the spam filters notice that an increasing number of emails coming from a particular IP are spammy in nature or being marked as spam or violating the guidelines, it would start decreasing its reputation. Emails coming from that IP are marked as spam if the reputation comes down below a threshold.
- Domain Reputation: When you send emails with your domain, lets say email@example.com, there is also a reputation associated with yourdomain.com and spam filters check for this reputation before classifying your emails.Like the IP reputation, the domain reputation also plays a role in classification of your email as spam or not.There is a high chance for your emails to end up in spam when more emails from your domain are marked as spam or if they have spammy content.
- Sender’s Reputation: Some Spam filters score your email address with a sending reputation and this depends on the number of emails that are getting engaged with (open, reply and bounce rates). When you create your email for the first time you have a neutral sending reputation and as you send emails to your prospects depending on how your prospect engages with them, your sending reputation increases or decreases.You need more of your prospects to engage with your emails for better sending reputation.If you are sending to unknown people it is always advisable to clean your list i.e validate your emails before you send emails because if there inaccuracies in your list it might bounce and a higher bounce rate also effects your sending reputation.Hence it is always advisable to warm-up your campaigns by sending fewer emails initially and slowly increase the number of emails you are sending.With a low sending score if you are sending more number of emails, then there is a higher chance of your emails getting throttled or ending up in spam
- Volume: If you are using G-suite accounts to send emails, though it officially says that you can send 2000 emails per day. But after watching millions of emails sent from gmail, we have noticed that there is a higher chance of your emails reaching inbox if you are sending to less than 500 emails from gmail. Sending under 500 emails with 40–50% open rates is way better than sending over 1000 emails and having 10–15% open rates.
- Having right personas on the List: One thing you need to make sure at the time of making your list is to ensure that the list consists of prospects who might be interested in your proposal. If your prospects don’t find your proposals as something they would be remotely interested they might mark you as spam. It is a big no in cold emailing. You should have sufficient info on your prospects background before reaching out to them.One other thing is that you should validate the email ids of the prospect. Sending emails to invalid email ids would result in higher bounce rate. Having a higher hard-bounce rate indicates that you are guessing the emails. This reduces your sending reputation because spam filters act against name-space mining.
- Authentication: Having SPF, DKIM authenticated and having your DMarc records updated is extremely important for spam filters to know that a person who is designated to send from that domain is actually sending email and it is not phishing email.
- Spam Complaint Rates: Spam filters have been most ruthless on this particular aspect in their quest to reduce spam. If you are sending cold emails, make sure that the information you are giving out is relevant to your receiver. If it is not, there is a high chance that they might mark you as spam. In order to reduce the number of complaints, give your prospects an option to opt-out / unsubscribe from your campaigns. And this should be of least resistance to them.Most spam filters have little leniency as far as spam complaints are concerned. For certain spam filters, getting more than 20 complaints for every 5000 emails your emails will end up in spam. This rate might differ for email clientsGmail has been sending the reason for a mail it sends in spam. If your email has been marked as spam by your prospects, then it says: “Why is this message in spam? Lots of messages from somesamplewebsite.com were identified as spam in the past.”
Having an opt-out /unsubscribe link should be a must if you are cold emailing.
- Content: It is important that you don’t go for templates from the web and use it for your campaigns. Because a lot many people might have already used it for their purpose and may have abused it already. Spam filters also matches the content. If it finds similar content which was marked as spam earlier then it will send your emails to spam with the message: “ Why is this message in spam? It is similar to messages that were identified as spam in the past.”I have personally tested a lot many use cases in this regard and have seen having certain keywords in your email or how using templates would send emails to spam.Also never use the same message for a very long time. It is always advisable to change your messaging from time to time.If you do all the above things right and spend a little extra care while doing cold emailing, you would reap much better results. After all getting results is more important than just sending more emails and ending up in spam.