How not to get people to open your emails

Rebuttal to Julian Shapiro’s TechCrunch article, “How to get people to open your emails”

Skyler Holobach

To the uninformed, email marketing seems simple. You send a message, someone at the other end receives it, and the transaction is over. In reality, sending mass email takes more careful thought, planning, and awareness than the author suggests.

The TechCrunch article this references has one major, fundamental flaw: it doesn’t understand what open rates are. In saying that every sender can get a 100% open rate, it fails to take into account just how inaccurate a metric open rate is.

Before we talk about the specifics of the article, let’s remind everyone what an open rate is. Email marketers put small 1x1 pixel images in an email. When that image is loaded, an open is recorded. Opens can be recorded before the recipient even gets the email in their inbox - some filters do this to check and make sure the images are safe. Recipients can open emails without ever loading an image. Many people block image loads on default, and some modern email clients (like apple watch) don’t load images ever. There are even some corporate filters that block any 1x1 pixel load, even though the end user sees other images inside the email.

Excerpts from the original article are quoted below, with responses below.

First, a few obvious pieces of advice for avoiding low open rates: Avoid spam filters by avoiding keywords commonly used in spam emails.

If it’s 2002, this is great advice. But it’s 2019, and spam filters have moved on from classifying mail as spam based on specific trigger words. There are hundreds, if not thousands, of characteristics a spam filter looks at to determine if a message is spam, and filtering is constantly evolving to keep up with spam. Email deliverability is not as simple as “avoid these words”, because spammers could also avoid those words just as easily, rendering trigger word advice effectively useless.

Further reading can be found here and here.

Consider using email subjects (1) that are clearly descriptive and (2) look like they were written by a friend. Then A/B your top choices.

Clear, descriptive subject lines are extremely important, not just for your subscribers but also to ensure you do not potentially violate the law. Writing subject lines that look like they were written by a friend can be dangerous territory, and is generally not recommended. Writing a subject line that says “Hey, are we still on for your wedding registry setup on Saturday at 2pm?”, which is something I might expect from a friend or my wedding coordinator, when you are actually a retailer advertising a sale is a quick way to get me to unsubscribe, complain about spam, and potentially complain to your ESP.

Include the recipient’s name in your email body. This signals to spam filters that you do in fact know the recipient.

Research conducted on those in the email industry found, on average, 38.5% of messages in the spam folder were personalized with the recipient’s name.

None of these supposedly personalized made it to the inbox–including my name had no effect on deliverability. It didn’t “signal to spam filters” that they knew me. Personalization is the minimum bar at this point for email sending- as a recipient, I expect my emails to be personalized (and not just with my name, but that’s another article for another time). To suggest personalization is something that improves deliverability is misinformed. Personalization may increase engagement if your mail is wanted in the first place, but it’s not what’s going to get you in the inbox.

Now, for the real advice: Let’s say 60% of your audience opens your mailing, how can you get the remaining 40% to open and read it too?

According to Mailchimp’s research, average open rates range from 15%-27%. Starting with a 60% open rate as your hypothetical is at best, ill-informed. At worst, it’s perpetuating the belief that open rate is anything more than a vanity metric and is something your team should be focusing on and striving for.

There are too many factors that affect the open rate metric that make it a poor way of judging the success of your email campaign. Most importantly, open rate calculations depend on image loads, which don’t occur if someone doesn’t have image loading turned on by default. Opens will also not be tracked if someone opens the text version of your email and doesn’t receive the HTML version.

Steve Atkins of Word to the Wise sums it up best with this quote: “If you’re a legitimate marketer – open rate has never been a KPI. [It’s] a useful metric but does not measure performance [of your campaign].” If you’re optimizing for open rates, you’re not actually paying attention to communicating with your recipients.

Important note: The reason many recipients don’t open your email is because it was sent to Spam, it was buried in Promotions, or it was insta-deleted because it looked like spam (but wasn’t). The goal here is to resuscitate these people.

This is not the main reason people don’t open your emails. There’s a good chance people aren’t interested in what you have to say. They’re too busy and your message isn’t that important. Focus on the people that ARE interested in hearing from you instead of the people that are not engaging.

A major contributor to emails not being delivered (or opened) is due to bad sending practices coupled with a low-engaged audience. If you are sending emails to those that do not want, or did not sign up to receive them it is highly likely ISP’s will catch on to this and filter your messages where they belong, in the spam folder. You can ensure your list is clean and engaged by only emailing contacts that explicitly opted in to receive your emails (this means no renting or buying lists, either).

Free inbox providers look at IP reputation and authentication (SPF, DKIM, DMARC), but engagement may be weighed just as heavily. These days, having a decent IP reputation and maintaining your authentication protocol may not be enough to get your emails to their intended recipients. Non-engagement filters also include signals such as “does this mail look unsolicited?” For example, if an organization is buying, harvesting, stealing, or otherwise acquiring lists in a non-permissioned way, then they will have high bounces, send mail to users who no longer (or never) existed, or generate complaints.

Let’s also stop perpetuating the belief that the Promotions tab is where email goes to die. Gmail is not the enemy. The Promotions tab is not the spam folder. If you are sending a promotional email, that is where it belongs. Play nice with Gmail, follow the rules, and you shall reap the rewards. (More reading here.)

(1) Duplicate the initial email then selectively re-send it to non-openers. This time, use a new subject (try a new hook) and downgrade the email to plain text: remove images and link tracking. De-enriching the email in this way can help bypass spam filters and the Promotions tab.

We get to my favorite part of the bad advice: Schrodinger’s Email. Remember the earlier part of this article, where I said open rate calculations depend on image loads? If you downgrade the entire email to all plain text, you will never know if that email was opened. You are also not more likely to bypass spam filters with an all text email, so this advice takes the cake for the worst advice in this article.

Email marketing looks simple on the surface, but, as evidenced by the email community’s response, it can be quite complex. Before acting on clickbait advice you find on the Internet, reach out to the Email Geek community on Twitter with the hashtag #emailgeeks. You’ll be sure to find a knowledgeable email person who is happy to help out!

Nout Boctor-Smith, Stephanie Griffith, Laura Atkins, and Steve Atkins contributed to the writing of this article.

Skyler Holobach manages Email Compliance for Marketing Cloud and Pardot, is the Women of Email Anti-Harassment Chair, frequently speaks at M3AAWG, loves winning arguments, and hates character counts.