Marketing’s Weekly Dose of the Truth

Ken Magill

About Us

Engagement Discussion Based on BS ... For Years

2/17/15
 
By Ken Magill
 
Information reportedly revealed during a panel discussion at the Email Experience Council’s conference in Florida earlier this month has turned the whole idea of email engagement on its head.
 
According to Dela Quist, chief executive of email marketing agency Alchemy Worx, John Scarrow, general manager of safety services for Microsoft, and other email inbox provider representatives said during the panel discussion that they do not monitor clicks to determine whether or not incoming messages should be treated as spam.
 
What’s more, according to Quist, Scarrow said as long as marketers follow list-hygiene best practices, it is unwise for them to remove email addresses from their files after a year or two of inactivity.
 
“The guy from Outlook.com said anyone who removes their inactives—and I’m not talking 10 years—but he said if you remove your one- and two-year inactives from your list you're leaving money on the table,” said Quist.
 
“They then went on to say: “We do not measure clicks ever as a measure of engagement,’” said Quist. “’It’s against privacy. Even if we could we wouldn’t.’”
 
“The guy said: ‘We kind of measure opens, but not in the way you guys think,’” said Quist. “’Engagement to us is moving into the spam folder, out of the spam folder, replying to an email. That’s engagement.
 
“And they also said that engagement in its own right will not get a sender blocked,” said Quist.
 
If Quist’s recollection is correct—and it has been verified by another attendee in a blog post I will link to below—it would mean one of the most prominent ongoing discussions in email marketing has been based on a load of nonsense.
 
Put another way, a lot of so-called experts in permission-based email marketing are completely full of crap and have been full of crap a very long time.
 
For years, many email service providers' deliverability experts have argued that ISPs make deliverability decisions based on—among other things—on recipients’ engagement, or lack thereof, with senders’ messages.
 
In defining engagement, they almost always cite clicks as a top metric.
 
ESP experts have also argued over the years that removing inactive addresses is a way to drive so-called engagement metrics up and improve the sender’s email reputation.
 
And now we hear the major ISPs don’t even look at clicks as a measure of engagement. And one representative even said a lack of engagement alone will not get a sender blocked.
 
This would mean that any marketer adhering to the advice of an expert telling them they should remove inactive addresses from their files to increase their engagement metrics such as clicks has been following some very bad advice.
 
Moreover, they have certainly lost sales as a result. Extrapolate the idea of marketers needlessly removing inactive addresses and losing sales as a result across the industry, and the lost revenue potential is staggering.
 
So-called experts have cited clicks as a primary engagement metric repeatedly.
 
Some examples:
 
“Engagement (a measure of the number of people who open your email, click on links versus the total number of people to whom the email was sent) is going to carry the most weight in the coming months,” said a 2013 CakeMail blog post.
 
“While open rates and click-throughs are typically considered engagement metrics for email marketing campaigns, Internet Service Providers (ISPs) look at them as a test of deliverability,” said a blog post from Lyris in 2012.
 
“From a deliverability perspective, click-through rate is the number one engagement measurement that ISPs rely on because it shows a definitive action taken by the email subscriber,” said a Responsys blog post in March.
 
“Many ISPs now look at customer activity to determine whether or not to deliver email. ISPs often use custom algorithms to measure engagement, but common metrics may include the following:
 
 
“2. Clicks,” said a 2014 SendGrid blog post.
 
“Engagement is a metric used to determine which customers are interacting – that is, opening and clicking – with your emails. ISPs such as Gmail know who is opening and clicking on your emails; they also know who is viewing emails, and who’s deleting before or after reading. They crunch this data to determine engagement rates,” said a 2013 Act-On blog post.
 
And here is an example of an expert recommending removing inactive addresses based on, among other things, a lack of clicks:
 
“Are your subscribers opening and clicking your emails? Or do they delete your emails without opening them, unsubscribe, or report them as spam? Gmail uses such engagement metrics to determine whether its users want your emails. So if your subscribers are inactive — i.e., if they haven’t opened or clicked your emails in a certain period of time, such as 6 months — you should consider running a reengagement campaign. In this way, you’ll identify the subscribers who aren’t interested in receiving your emails and can remove them from your list,” said a 2012 blog post on FulcrumTech.
 
Does this development mean ISPs don’t look at engagement? No. But they reportedly look at different metrics than many claimed.
 
According to a report on the EEC event posted on the MailUp blog by Massimo Arrigoni, in order to assess engagement, ISPs look at opens, replies, “move to junk,” “not junk,” “delete without open,” “move to folder,” and “add to address book.”
 
The post is a must-read for Magill Report readers.
 
Note the absence of clicks as an engagement metric ISPs reportedly track. And remember the Microsoft representative said they track opens “but not in the way you guys think,” according to Quist.
 
The metrics the ISPs do reportedly track are not available to marketers unless they employ one of the email intelligence vendors who claim to be able to track such metrics with their panels, such as Return Path and eDataSource. [Full disclosure: Both are clients.]
 
In any case, a lot of people have some ‘splainin’ to do.
 
Comments

Show: Newest | Oldest

Post a Comment
Your Name:
Subject:
Comments:
Verification:
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: Dela Quist
Date: 2016-08-24 03:44:30
Subject: ESP's Clicks and engagement

Hi Ken

Good article as always yesterday on BS masquerading as best practice - looking back through the various times you have written about it I came across a quote I gave you 4 years ago and 3 years before this article. In it I said
“Categorically, they [ESP's] 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,” http://bit.ly/2c5wiJR Plus ca change
Posted by: Parry Malm
Date: 2015-02-23 10:33:41
Subject: Wowzers

If you don't send out sh*t and if you don't collect dodgy addresses you ain't got no problem. It's amazing that there's trolls even within the email geek community. Haha. I dream of a magical land where people just worry about sending out emails that make money.
Posted by: William Clinton
Date: 2015-02-20 14:42:00
Subject:

I did NOT inhale this crap! But hey, thanks for feeding the trolls Magill.
Posted by: Brian Williams
Date: 2015-02-20 03:05:23
Subject:

Don't let the haters get you down, Ken. I'm with you, my friend,.
Posted by: Deliverability Douchebag
Date: 2015-02-20 02:35:27
Subject: DO YOU EVEN REPORT, BRO?

A more responsible writer might try to get a comment on the record from a primary source instead of relying on one man's recollection of a few minutes of the event - especially when his recollection so conveniently supports his own long-held and frequently debunked opinion. Why not ask Mr. Scarrow himself? Maybe a comment from the EEC organizers? Nothing in the piece indicates that our intrepid reporter even attempted to obtain comment from a primary source before publication. Another ace job, Ken. Maybe your business model is not your problem.
Posted by: Dela Quist
Date: 2015-02-19 18:53:27
Subject: There was another half.....

Ryan I think you are being somewhat disingenuous when you imply that my points are invalid because I left halfway through the debate. To use your analogy you don't have to watch the whole of "Black Hawk Down" to know what happened in the end. Moreover nothing you or anyone else has said gives any indication that there was an alternative ending to the one I highlight. Deliverability and list management best practice has never been in question as far as I am concerned and in case anyone isn’t aware of this you and Dennis are people I truly respect and hang out with the most in this industry and for the record I was really impressed that you were able to put together such a distinguished panel to help shed light on engagement arguably one of the most vexing of questions to face email marketers. What do ISP’s mean when they say they measure engagement and how does that impact inbox placement. This issue is not a new one I first wrote about it in 2010 btw http://bit.ly/1zk7sFn and by way of reminder for you and for those who were unlucky to have missed the movie let me remind you that in the preamble to my question I pointed out to the panel and the audience that because the ISP’s have never stated for the record what engagement they actually measure there has been a knowledge vacuum. Unfortunately some people whether they are aware of it or not have taken advantage of this vacuum to pass all responsibility for deliverability on to the client. Now I know for a fact that this is not the case and that is what Ken’s article is about - he is right to call BS on those who have spent the last 5 years spreading this kind of disinformation. So find it a little curious that you don't even acknowledge that it was my question that turned what would have been a routine and dull panel about deliverability into one of the most talked about panels in the history of the EEC. Why? Because John Scarrow and the other ISP’s – it wasn’t just Outlook.com btw - FINALLY gave us a sense of what they measured! Now believe you me I am equally concerned that John Scarrow is being hung out to dry over this and I would like to say for the record that he was the most honest and outspoken person on the panel particularly when it came to inactives - the issue that seems to vex you so much. I know Sri Somanchi talked of ramp-up and ramp-down to 3 – 6 month inactives, but I am going to call BS on him. Sri I challenge you to provide me or Ken with empirical evidence that a respected brand sending email to an opt-in list can increase engagement as Gmail measures it and therefore inbox delivery by ramping down their send frequency to 3 – 6 month non openers. Now I am not talking rates here, for the reasons John Caldwell has identified. I am talking totals – both for engagement AND inbox placement. Sri if you really do measure engagement on a per subscriber basis, tell me how can they be more engaged with an email I don’t send and they don’t get? Ryan to your final point that “there are so many things that we can do in email. If a marketer wants to focus more on the 65% of records that are inactive, then so be it. There is a lack of sophistication on strategies that focus on those that are not engaged than those that are.” REALLY! Firstly if that number is 65% then you may need help, typically the number of inactives on a list as defined by opens or clicks is generally between 35% and 65% we generally tell our clients to set a 12 month benchmark of 40% - 45%. In other words they should aim to 55% - 60% of their list should have opened at least 1 email in the last 12 months. We call that open reach http://bit.ly/182a83R and we advise them to do so for revenue optimization reasons and not inbox delivery. If you want to talk sophistication, there is nothing less sophisticated than giving up trying to get non openers to open.
Posted by: Dela Quist
Date: 2015-02-19 16:19:44
Subject: There's always a nuance aka DBS

John I agree 100% as you know it's what I call the Don't Be Stupid rule. I always assume my audience is intelligent - many guru's assume the opposite
Posted by: Dela Quist
Date: 2015-02-19 16:05:10
Subject: On a slight tangent

Andrew - I would not use this debate as a reason to change your IP warming strategy. YOUR openers and clickers ARE engaged. The question is whether ISP's consider YOUR non openers engaged or not
Posted by: Len Shneyder
Date: 2015-02-18 18:10:18
Subject: re: On a slight tangent

I think its still applicable... if ISPs, and let's face it, we're really talking about Outlook, takes not of when someone deletes an email without opening it, that's a signal of 'disengagement' the absence of engagement is as important as the act thereof. If that's the case then warming up IPs to your most engaged audience improves the likelihood that someone won't delete your email without opening thus it stands to reason that you build a positive reputation on your IP.
Posted by: Andrew McIver
Date: 2015-02-18 17:49:10
Subject: On a slight tangent

Would this have implications for IP warming strategy? Standard practice is to ramp up volume to the most engaged cohort to get good engagement metrics on the initial campaigns. If engagement metrics aren't considered by the ISPs, does that make volume the only consideration – assuming list hygiene isn't an issue – when warming an IP?
Posted by: Ryan Phelan
Date: 2015-02-18 17:37:32
Subject: There was another half....

Interesting article, but let me lay out a few things. While Dela did hear what he heard, he left with 45 minutes left in the conversation. It's like sitting through a movie and thinking you know the end. First, many will fully admit that deliverability is not a "known" science. For years, methods that we have employed and recommended to clients were based on a educated guess. We don't need to "splain" anything, we have to do what we have always done. Adapt to new information. Microsoft made a statement as it applies to their domain, but Microsoft is not the entirety of the ISP/Receiver community. AND, it's not the only thing that was said. Walking out after hearing that one comment is listening for what you want to hear, not what was said. Second, the term clicks and engagement have been interchange for years. What we know is that a true defined “click” which is on a URL the ISP’s will not track due to privacy issues and well, those don’t go through their systems. engagement, is the things ISP’s do see now like the historical move of email, interaction, and read status. From the affirmative actions to the actions that are tagged as negative. To say that engagement does not exist is a mis-perception of the conversation. There were numerous quotes around the complexity of the algorithms that are utilized by the receivers. Gmail's represenative disagreed with Outlook in actually changing cadence DOWN. So, let's be a little cautious in saying that everything changed because of one comment. Third, we heard that there are 2 measurements of Spam. The personal and the domain. Personal is using those things that are engagement behaviors. When you have enough personal engagement actions on a sender, then you have implications on the domain level. Lastly, there are so many things that we can do in email. If a marketer wants to focus more on the 65% of records that are inactive, then so be it. There is a lack of sophistication on strategies that focus on those that are not engaged than those that are. The use of data is where we as marketers need to move as an industry because we are dreadfully behind in our sophistication. Anyone can send more email and make more, but they can also burn out a list. I've seen it done with a "send more" philosophy. let's try to focus on a deeper level of conversation that involves data sciences, segmentation and personalization to the consumer instead of just throwing up all over those that like/love our brand. Yes, the panel changed the thinking of our approaches and for one, I embrace the change, but let's not simply chalk up what the kind man from MSN said to apply to all email practices
Posted by: John Caldwell
Date: 2015-02-18 14:51:41
Subject: There's always nuance

I think some of the moving parts are too easily convoluted in sound bite conversations. For example, removing inactive addresses will drive metrics up in that if you have a based of 500K recipients with a 20% open or click rate and you remove an inactive 10%, or 50k records, you automagically increase your open or click rate to 22.5%. If you're a marketer under pressure to show metric improvement you just did. If your ESP costs are CPM-based and 50% of your 500K base has been inactive some time, cutting the dead weight could reduce your CPM costs, and operational savings go straight to the bottom line profit. If you're a marketer under pressure to cut costs you just did. Of course, taking either at face value is as ridiculous as taking "send more email" at face value. None of the above - or "sending more" - comes without consequences so none should come without careful consideration. A successful career in email marketing doesn't come from checking the right boxes because this that or the other person says so, which should be evident by this article....
Posted by: Dela Quist
Date: 2015-02-18 11:57:33
Subject: Forest for the trees

Smartypants Your points make me think you did not actually read the article
Posted by: Dela Quist
Date: 2015-02-18 11:51:11
Subject: Make Use of What We Got

Hi Anthony Thanks for your contribution and more importantly going on the record. The point is they are not dead weight. Firstly iSP'S consider them to be engaged by the metrics they measure but more importantly based on sales data we measure from "long-term" non openers and clickers in other channels for our clients, not only is the revenue from this group significant it is also directly correlated to the number of emails they get, so someone who gets 2 emails/week spends more than someone who gets 1 email/week even if they both do not open a single one of the emails they get. I wish more people looked at the data that mattered. Dela
Posted by: Anthony Chiulli - Salesforce
Date: 2015-02-18 11:25:40
Subject: Make Use of What We Got

While I do agree with some of your points here Ken, I would argue that Marketers do not have access to many of the metrics or variables that ISPs have claimed they take into consideration. Adding the sender to your address book, moving out of spam, replying to an email are not consistent or quantifiable metrics in measuring engagement that a Marketer can make decisions on. Opens and Clicks are the best data points that Marketers and ESPs have at our advantage to identify delivery issues and engagement. I would also argue that even if Outlook mentions that Marketers are leaving money on the table by excluding subscribers who have not registered an open in over a year or 2 years, seems like dead weight if you ask me. How likely is a subscriber to buy when they haven't opened an email in years? Thank your for article, lots of great discussion from this EEC meeting.
Posted by: Smartypants
Date: 2015-02-17 19:48:11
Subject: Forrest through the trees

Outlook.com does not equal all ISPs. And even if it did, using a set of known, quantifiable metrics which are highly correlated to a set of vital but unquantifiable metrics isn't really all that stupid. Dela needs to do some 'splaining as to why he is singlehandedly trying to opens and clicks as meaningful metrics. They don't mean everything, but they are critical metrics when measuring user interest as well as insight into user engagement.
Posted by: Steve Mintz
Date: 2015-02-17 18:16:49
Subject: Experts

Thanks for sharing, Ken. I will be very curious to see comments to your post from some of those same deliverability experts 'splainin'. Perhaps you can ask some of those cited in the posts above some basic questions, kind of like you did with Spamhaus last year.

Xverify