Marketing’s Weekly Dose of the Truth

Ken Magill

About Us

ISPs Judge You on a Metric You Can't Measure


By Ken Magill

Forget individual opens and clicks.

In their efforts to block email people don’t want and deliver wanted messages, Internet email inbox providers are reportedly assessing emailers’ spam reputations based on something called the “read rate”—the percentage of a sender’s messages that get read.

So what does a marketer’s read rate need to be in order to avoid having their messages being treated as spam?

If ISPs have anything to say about it, we’re never going to know. The read rate is based on information only the ISPs have. If they were to publish what they consider an acceptable read rate, spammers would most assuredly game it as they do everything else.

As a result, ISPs are assessing email marketers on a metric they can’t measure.

According to Laura Atkins, principal at email deliverability consultancy Word to the Wise, the so-called read rate isn’t new. It’s also a term used by marketers, not ISPs, she said.

“The big ISPs that control the interface (webmail and AOL) and/or have access to IMAP [Internet Message Access Protocol] data have been using this measurement for a while,” she wrote in an email exchange with The Magill Report.

“It's really looking at what emails are opened by the recipient. The real reason it's called ‘read rate’ is to distinguish it from the marketer measurement ‘open rate,’” she added.

An “open” is recorded when the receiving machine calls for graphics from the sender. As a result, it’s a flawed measure, but it’s the closest tool marketers have at their disposal for determining how much of their email is getting opened.

Atkins explained further:

“Read rate is a real measure. It's who opened the message as measured by flag changes in the mailbox / on the spool,” she wrote.

“Open rate is a proxy. It's who displayed images in the message.

“Read rate can be measured by people who have access to the user's inbox. So the ISPs and OtherInbox [a company recently acquired by deliverability firm Return Path] can measure it. Senders, ESPs and 3rd parties cannot measure it. Without access to the user's inbox no one can get this number.

“Open rate can be measured by people who have access to the server images are hosted on. So senders and ESPs and 3rd parties can measure it. ISPs cannot. But this requires the cooperation of the recipient - they have to have images on,” she continued.

“Read rate may be higher than Open rate. Open rate is never going to be higher than read rate,” Atkins wrote.

That ISPs are using the so-called read rate—even if they don’t call it that—to assess marketers’ email campaigns runs counter to the idea that they look at opens and clicks.

“Categorically, they are not looking at open rates as we measure them and as I understand it don’t look at clicks at all, and anyone who says they are doesn’t know what they’re talking about,” said Dela Quist, CEO of email marketing agency Alchemy Worx. However, he said: “Open rates can closely correlate to read rates.”

Also, Quist said, worrying about read rates and the fact that senders can’t measure them is a waste of time.

“It’s pointless,” Quist said. “We might as well just focus on being a good brand with good practices for acquiring names. Deliverability troubles start at the point of acquiring names and they end there.”

Indeed, whenever an email deliverability specialist has a client who has been blacklisted by anti-spam groups like Spamhaus, generally the first thing they look at is how the client has been acquiring email addresses. For example, buying email addresses is usually a fast track to getting blacklisted.

Also, Quinn Jalli, senior vice president, strategic initiatives group at marketing services provider Epsilon, contends that the open rate, as flawed as it is, is a perfectly fine metric for determining if a marketer’s messages will experience delivery troubles.

“What I can tell you, unequivocally, is that any client of ours who has a 20 percent open rate has no deliverability issues,” said Jalli. “Three percent or less and you’re in trouble. At five to seven, [percent] you’re in that gray area where you’ll probably have troubles at Gmail and you may have troubles at Hotmail.

“Then you go up to 10 to 15 [percent] where you may have a little trouble at Hotmail—because you’re hitting a hard filter, but that’s not likely—and maybe a little trouble at Gmail, but overall you’re in a pretty good place.

“And above 15 percent, you’re across–the-board golden,” Jalli said.


Show: Newest | Oldest

Post a Comment
Your Name:
Please type the letters in the image above

Terms: Feel free to be as big a jerk as you want, but don't attack anyone other than me personally. And don't criticize people or companies other than me anonymously. Got something crappy to say? Say it under your real name. Anonymous potshots and personal attacks aimed at me, however, are fine.

Posted by: Andrew Barrett
Date: 2012-08-28 21:31:30
Subject: Shaking angry fists at e-mail gods

Complaining that ISPs do not share metrics with senders is a bit like howling at the moon: interesting, possibly, for a short time, but is manifestly ineffective and quickly becomes tiresome. We, as senders, are not the ISPs' customers - in fact, we use their infrastructure for free. Expecting a good customer experience from the ISP is silly. We are not their customers. To complain that ISPs will never share read rate thresholds with senders is old hat. They are already giving us more data than we pay for - and in many cases, more than we actually use. Some even give us a top level read on IP reputation and spam trap hits. It would be nice to get more to go on, but that really should be enough for smart senders to get a bearing on where their problems lay.
Posted by: Mickey Chandler
Date: 2012-08-28 16:05:57
Subject: Just to expand on my last comment a little

Most of the SBL cases that I've dealt with recently have involved people who have been mailing old lists where they've been trying to "keep our name in front of people." The issue was that some number of those "people" no longer had their old addresses, and those addresses had since become spam traps. Since these clients had made it too hard to leave their lists, they had lots of really old addresses. And that incrementally increased the chances of getting listed until it finally happened.
Posted by: Mickey Chandler
Date: 2012-08-28 15:55:36
Subject: And they end there?

I'm not entirely sure where Dela Quist learns where deliverability troubles start or end but it's not from anyone that knows a single thing about how the real world works. I deal with Spamhaus SBL issues quite extensively. And I do ask a lot of questions. And yes, those question start with "How do addresses get onto your list?" But, the questions very rarely end there. I usually have to go on to the other set of painful questions: "How old are these addresses?" and "How hard is it for an address to leave your list?" That means things like unsubscribes and removing inactives. But, you know, when you've got a hammer...