Email Newsletters Hit a Speed Bump: Inaccurate Open Rates
Email newsletters are hot, with whole digital-media startups built on their backs and traditional publishers like The New York Times and The Washington Post producing them by the dozens.
The conversational news roundup TheSkimm, for example, says it has over three million subscribers. Lena Dunham and Jenni Konner's Lenny Letter claims hundreds of thousands. That reach is valuable when drawing readers to websites' homepages keeps getting harder. Both are now focused on building sustainable businesses, backed by venture capital at TheSkimm and by magazine powerhouse Hearst in the case of Lenny.
But another newsletter flagged an unexpected challenge last week, one that caused its founders to rethink their primary method of distribution and introduce an app to complement it. Liza Darwin and Casey Lewis, who started Clover Letter in February for teen girls, wrote on Medium that their strong start was running into a problem with their open rate.
"A week into launch, we had a couple thousand subscribers and a 70% open rate," they wrote, under the headline "The Problem With Email Newsletters." As they got more comfortable with MailChimp, the service they used to manage and send Clover Letter, they said they began adding "GIFs and Instagrams and other things that made the experience feel that much more exciting than your average e-blast. And that's when our open rate began to plummet."
But, the two wrote, the drop wasn't because readers were actually opening Clover Letter less often. They traced the issue to the way MailChimp determines whether a newsletter has been opened, which relies on a "tiny, transparent image" that is loaded when an email has been opened. (MailChimp explains the process on its website.)
Large newsletters can get "clipped" by email programs, which hurts the open rate because the image is at the bottom of the email. Gmail, for example, clips emails that are larger than 102kb and puts everything beyond that mark behind a link that says "View entire message." Readers who open the email and read much of it but don't click that link will never give the tracking image a chance to work.
Offline reading, when an already loaded email is opened while a reader doesn't have an internet connection, can also contribute to the problem.
For both editorial and commercial purposes, the "open rate" metric is important. Editors might use it to help gauge whether their subject lines resonate. For potential advertisers, it suggests something about a newsletter's actual audience. Nicholas Quah, who writes a newsletter about the podcasting industry, compared open rates to websites' click-through rates, as far as advertisers are concerned.
Clover Letter explored other newsletter providers, but "came crawling back to MailChimp," Ms. Darwin and Ms. Lewis wrote. Looking for alternatives, Clover has created an iOS app, though it's not doing away with its namesake newsletter.
MailChimp did not respond to requests for comment. Its website acknowledges that its open rate isn't 100% accurate and urges publishers getting clipped by Gmail to keep their newsletters below 102kb.
Publishers may not embrace that solution.
"We're talking to developers and weighing options, but based on reader feedback, abbreviating the letter would be doing a disservice to the audience," Ms. Lewis told Ad Age in an email.
Ms. Lewis said Clover is focused on growth rather than brand deals at the moment. Dove served as Clover's launch sponsor, but the newsletter doesn't run traditional advertising. "It's unclear just how aware advertisers are that open rates can be inaccurate because it seems like no one really talks about it," she said.
Benjamin Cooley, CEO of Lenny, said spam filters used by providers can also prevent readers from getting their newsletters.
"As the Clover creators noted, a lot of the problems with delivering email is that Outlook, Gmail, and other email servers look for certain words and tags in your email to determine if they are friend or foe," he said. "When emails become laden with a lot of pictures and gifs or use words like 'vagina' one time too many (as they sometimes will when talking about women's health) the email gets blocked."
Jacqueline Boltik, who helped Lenny get up and running, said she doesn't think the newsletter's length has had an impact on its open rate. Though, she noted, a concerted effort was made to keep the design code "lean."
She mentioned another potential measurement bugaboo: Some desktop readers only view the newsletter in Microsoft Outlook's preview feature, which wouldn't count toward the open rate.
Mr. Quah said he's seen his open rate "abnormally drop" when editions of his newsletter are clipped, but guessed that, overall, his metrics are only off between 5 and 10 percentage points.
Mr. Cooley, who admitted that he has a lot to learn about how newsletters work ("What I don't know or understand about how newsletters actually work could fill volumes"), summed up the conundrum.
"We are only as strong as our weakest link and sometimes the technology that enabled the revolution turns on us," he said.