Email marketing can be downright scary. Unlike some other forms of communication, there is no “Undo” button in the world of email. Fear can easily set in when you are ready to hit that big red “Send” button to your subscribers. Are there any typos, broken links, etc.?
Outside of the big red “Send” button, email marketers have other concerns that keep them up at night. Whether it be hitting spam traps, ending up on a blocklist, or going to the spam folder—simply getting your emails into your subscriber’s inboxes can be a scary task.
With Halloween around the corner, we wanted to take a deeper dive into 3 different topics that send shivers down email marketers’ spines.
How can we ensure we do not spook the email filters with our Halloween campaigns?
Halloween is that time of year when you go costume shopping and pumpkin hunting. But trick-or-treating online is not as fun, especially for the mailbox providers (MBPs).
When Halloween comes around, MBPs begin to see senders ramp up volume. This has been offset by Amazon Prime Day in June, but nevertheless, this is the time of year when the filters anticipate higher volumes.
Look for the following 3 things to avoid MBPs haunting your dreams.
Filters love patterns, and having a consistent email volume is the first step toward building a pattern. As a sender, you might want to try new campaigns and offers around the holiday season but not before doing a proper warmup.
If you do anticipate higher volumes, an additional IP address is the way to go. A good rule of thumb is to check if your sending volume will increase more than 2x in the coming weeks compared to your current volume. Keep in mind that spammers also warm up IPs but usually send as much as possible before moving to a new one.
As a sender, however, we do not want to exhibit any of those behaviors. When MBPs see higher volumes, the first response is to throttle email streams. If you feel your email volume is already high, please check with your Email Platform Manager for additional IPs so that your email delivery does not slow down.
It is important to keep in mind that adding the IP too late into the holiday season can be counterproductive. MBPs look at the history of the IP, which can take more than 30 days to build a reputation. This is why you should establish the reputation of the IPs and domains/subdomains much in advance of your large sending season.
Testing and monitoring
Now that you are all set with your Halloween creatives and can’t wait to unleash your creativity to your audience, hold on to it—you need to test your email and examine the metrics closely. Are the filters being picky about the content? Is your authentication and custom URL tracking working as designed? (Hint: Did you know domain reputation is built based on your DomainKeys Identified Mail (DKIM) domain?)
Once everything looks good, send a small campaign to your most engaged users and build trust for the campaign before going deep into your database, pulling those dead email addresses from the grave. Take it slow and try to space out your campaign as much as possible. There might be time-sensitive offers, but again, that is a common spammer tactic of showing a false sense of urgency. On any given day, more than 80% of traffic that MBPs see is spam, and as a sender, it is imperative that you stand out from the crowd.
By A/B testing your campaigns, you ensure that you not only stand out from the crowd but from other emails in your recipients’ inboxes. With the Postmasters taking a break during the holiday season, do not expect any quick fixes at the last minute. MBPs have a moratorium in place around the holidays that would make it difficult to push inboxing/reputation changes.
Contributor: Sri Chandran (Sr. Email Deliverability Consultant)
The open rate is dead
Email marketers did not need to wait until Halloween for a good scare this year. After releasing the iOS 15 on September 20th, Apple introduced Mail Privacy Protection (MPP). When a user has MPP enabled, the Apple Mail application fetches the message (and all the images in it), causing the tracking pixel to fire. Since the tracking pixel fires, an “open” will be reported, even though the user possibly has not seen the email yet.
When the news broke of the release of this new feature, email marketers everywhere went into panic mode. MPP has been a topic that has caused many sleepless nights for email marketers. Will this kill the open? How will I know who engages with my email? How will this impact my sunset levels?
Twilio SendGrid has released a full guide on MPP to answer all of those questions. The guide covers Apple’s new feature from all angles: What exactly is MPP? What does the data look like so far? How can senders adapt?
It has now been 3 weeks since the release of MPP. Is the open rate “dead”? Looking at the first 3 weeks since Apple released MPP, we can see 9.2% of total opens occur in MPP-protected mailboxes. You can also continue to rely on your open events that were not triggered by Apple machines, thanks to a new field with “Open Events” in our Event Webhook to identify “nonhuman” opens. More information on this new field can be found here.
However, some adjustments will be necessary as senders lose access to some open data. As a Deliverability Consultant, I have seen customers handle MPP in various ways. Some of the methods we promote are updating sunset levels that factor in machine open data, improving automation, and approaching reengagement campaigns using machine open data vs. nonmachine open data.
Contributor: Denis O’Sullivan (Sr. Email Deliverability Consultant)
What Brendan Fraser and email have in common
No matter who or what messes with a mummy, that mummy will wreak havoc.
All jokes aside, as we approach Halloween and the less spooky holidays this year, it is important that senders do not give in to the temptation to bombard unengaged consumers (“mummies”) with email. While it may just seem like a scary campfire warning, the saying “Sending an email to anyone who has not engaged in 6 months is the kiss of death for your email program” is true more often than not. (Plus, that quote came from Microsoft’s anti-spam team, who we tend to listen to.)
Recently, we had a client who found themselves blocklisted by Spamhaus. With a stellar engagement rate throughout the past year on almost every campaign, we were surprised to see them listed where we typically see some of the worst senders. Upon further investigation, it turns out, the client decided to go from targeting 90-day engagers to its entire database for a holiday campaign. Within 24 hours, there were serious repercussions (blocklisted, deferred, and ultimately, blocked by many ISPs). The cause? A bunch of “mummies” hanging out in its database, resulting in the email hitting spam traps as well as dormant mailboxes.
While that was a one-off scenario that we were able to quickly resolve by reverting to the client’s normal targeting, this may not be the case for all senders. If you decide to target recipients who have not engaged in the last 6 months—especially if you plan to send them multiple holiday campaigns—the results could potentially be apocalyptic for your email program.
How to avoid database mummification
If there is a business use case for targeting a mummified cohort of your database, then here are some precautions you can take to minimize the negative impact:
- Do a test campaign that includes recipients who have not engaged in 6 months but have engaged in the past 9 months. If that performs fine, then you could do another campaign targeting those that have engaged in 12 months. Then use that to gauge how much wiggle room you have to extend your audience segments for the rest of your holiday sends.
- However, if neither of the 9 or 12-month tests performs well, that is your sign that you should not target that old of a cohort.
- Past 12 months, we cannot, in good faith, even recommend sending a test campaign.
- Send campaigns only to unengaged email recipients that also have a recent history of interacting with your brand on other channels (e.g., being active on your website and/or making a purchase).
- Generally, we recommend sticking with consumers that have had some sort of brand interaction within the past 12 months, preferably within the last 6 months.
- Consider waiting at least an hour after sending campaigns to highly engaged recipients before sending any test/live campaigns to your mummies.
Note that any testing should commence at least a week before your actual holiday campaigns go live. This should give you time to evaluate impact, mitigate any issues, and make adjustments/conduct additional testing.
At the end of the day, though, the only mummy incorporation we recommend is in your content images.
Contributor: Emily Thrasher (Sr. Email Deliverability Consultant)
The world of email marketing is ever-changing, and senders have to continuously adapt. Twilio SendGrid’s Apple Mail Machine Open Indicator, new guidance on domain/IP warmups, and fluid sunset levels are just 3 of those examples. If your team is looking for assistance in navigating this ever-changing world, our Deliverability Consultants are ready to help.