So the iOS 15 changes are out in the wild and I thought it was best to start a fresh thread to give everyone a place to ask questions, help each other, and share their experiences.
Some thoughts on how to think about Suppression in an Apple Mail Privacy Protection world.
First off, thanks for all of your comments, questions and participation in this thread. We appreciate your input and feedback about this change from Apple.
In 2016 we added Sends Since Last Engagement (SSLE) and its related feature, Graymail Suppression, in an attempt to make a task that had been very difficult for most folks, that is list hygiene, much easier. While neither feature was perfect, they did accomplish what we set out to do.
As you might expect, Apple’s Mail Privacy Protection has changed the way these features work, but probably not in the way you expect. In this post I’m going to share some more details on how things work and what changes we’re going to make in the product.
A little bit about Bot Filtering.
There are security vendors who primarily play in the enterprise space that have been doing this “open every email and click every link” thing for a while now. In fact, we’ve had an option to filter these so called “Bot Opens” in your portal’s settings since 2017 and it’s been defaulted to on for as long as I’ve known (see: Tracking & Analytics > Advanced Tracking and find the checkbox for “Bot filtering”).
Well as it turns out, our system is good, maybe too good, and it is correctly categorizing Apple iOS 15 opens as if they were from bots (because they are) and filtering them out automatically if you have this feature enabled. Note: for Apple iOS 15 events, this does not filter clicks, unsubscribes, or spam reports, only opens. While this prevents all the “false” opens from skewing your results, it also prevents you from having complete results.
Here’s a simple example: Presently about 10% of the traffic we’re seeing is being opened with Apple iOS 15 devices. So if you send to 1,000 recipients and have Bot Filtering enabled, you’ll only see opens from about 900 recipients - all of the folks who are not using Apple iOS 15 devices with Mail Privacy Protection enabled. But we do not actively remove sent events, so in effect, your open rate, in this example will actually appear to go down over time, though we believe the true activity has not changed.
So what happens if you enable check “Suppress my unengaged contacts” when you send?
If you have Bot Filtering enabled and suppress unengaged contacts, our system will not know if the contacts opening on an iOS 15 device are actually opening your mail. In this case, after 14 messages, if there are no clicks or opens on other device types, and if this Bot filtering checkbox remains checked, our system will suppress contacts that may actually be engaging with Apple iOS 15 devices.
What if you don’t enable Bot Filtering in your portal?
If Bot Filtering is disabled in your HubSpot portal, our systems will report any open that we see from Apple’s Mail Privacy Protection system as a normal open. The challenge in this case is that we will be unable to distinguish between a “true” open where the person actually opened your email and a “false” open where Apple is doing the opening on their behalf and then serving up the content from Apple’s own servers if and when that person decides to open the email. In either case, the events look exactly the same, though there do seem to be some edge cases that I won’t exhaustively go into here.
So what should you do?
This is honestly a tough question to answer and it depends on what you’re trying to optimize for. Each has risks and potential unintended consequences. Proceed with caution and care.
If you’re trying to optimize for list size and are not worried about triggering automations with lists filled with contacts that are added due to opens that might not be real, then you can periodically shut off Bot Filtering and deselect Suppress unengaged contacts when you’re emailing. This will allow you to recapture those Apple iOS 15 opens for any send that occurs when the bot filtering setting is disabled. So when Apple does their Mail Privacy Protection thing, you’ll receive the opens for contacts who would have otherwise been captured by our bot filtering and shown as unengaged. But keep in mind, you will get all of the Apple folks who are not actually opening too. As I said above, proceed with caution!
If you really want to make sure that you have very high confidence that those opens are truly opens, leave your Bot Filtering on and continue to suppress your unengaged contacts but really drive towards clicks as a new form of engagement. Find reasons for anything from social likes to upvotes, or article views or regular old CTAs, but get those clicks! Even if they are on iOS 15 with Mail Privacy Protection, we’ll know there’s an open when we see the click.
What is HubSpot going to do?
We’re continuing to work on filtering and identification systems for bots and other non-human opens. We’re also working on different reporting filtering options to enable you to better understand what is happening with your data. We understand that right now this is not easy and we are working to make it better. Finally, we’re working on other ways to measure engagement that are not just opens. While opens were certainly great, their time has passed. They’ve long since been a directional metric at best, but now, they’re really just not all that accurate.
@CHeckel welcome to the Community! I didn't see any new update, but I'm continuing to encourage clients to find other metrics to measure. Clicks, replies, meetings... more meaningful and measurable than just opens.
Hope you find a solution!
Did my answer help? Please "mark as a solution" to help others find answers. Plus I really appreciate it!
We just happened to notice this same thing and if you look at the timeline of the rollout of the new bot filtering HubSpot (July), we had a massive drop in open rates but click rates staying in the same range that they were. Bot filtering was turned on automatically for all accounts when they rolled it out.
If you see a drop in your open rates, I would recommend connecting with HubSpot Technical Support, as Support is included in your subscription, and they will be able to provide real-time assistance for this matter, where you are able to share screenshots and further information about this.
@CHeckel welcome to the Community! I didn't see any new update, but I'm continuing to encourage clients to find other metrics to measure. Clicks, replies, meetings... more meaningful and measurable than just opens.
Hope you find a solution!
Did my answer help? Please "mark as a solution" to help others find answers. Plus I really appreciate it!
Is it really bad? and why do I feel uneducated since I am not as worried. Someone please educate me. For the sake of argument, I can pull the numbers of quality leads I have and if the ratio of android users to iOS is'nt that much of a gap, then it's really not the end of email marketing, plus add to it, the number of iOS user that you have on that "list" of yours, what's the percentage of people whos actually knowledgeable to activate or opt in to that bot? --unless I'm being stupid thinking it's Opt - IN;Opt - Out situation. What's the worse can happen, this might just be a good help filtering who to contact, I'm sure, we'll figure it out, 1 way or another, lastly , I think as long we don't forget that there's still data that we can always check and see then we can always come up with a problem based solutions.
We would love to use 'email open - read' as this seems like a more accurate way of measuring if someone has actually opened an email. You only seem to be able to do this for individual emails which is frustrating when wanting to analyse multiple emails across entire campaign. It would be great if in camaping analytics 'influenced contacts' took the 'open-read' metric rather then opens.
Has anyone else started seeing drops in Open Rates, per what @TomMonaghan posted about Hubspot bot filtering?
****************
A little bit about Bot Filtering.
There are security vendors who primarily play in the enterprise space that have been doing this “open every email and click every link” thing for a while now. In fact, we’ve had an option to filter these so called “Bot Opens” in your portal’s settings since 2017 and it’s been defaulted to on for as long as I’ve known (see: Tracking & Analytics > Advanced Tracking and find the checkbox for “Bot filtering”).
Well as it turns out, our system is good, maybe too good, and it is correctly categorizing Apple iOS 15 opens as if they were from bots (because they are) and filtering them out automatically if you have this feature enabled. Note: for Apple iOS 15 events, this does not filter clicks, unsubscribes, or spam reports, only opens. While this prevents all the “false” opens from skewing your results, it also prevents you from having complete results.
Here’s a simple example: Presently about 10% of the traffic we’re seeing is being opened with Apple iOS 15 devices. So if you send to 1,000 recipients and have Bot Filtering enabled, you’ll only see opens from about 900 recipients - all of the folks who are not using Apple iOS 15 devices with Mail Privacy Protection enabled. But we do not actively remove sent events, so in effect, your open rate, in this example will actually appear to go down over time, though we believe the true activity has not changed.
How about making Hubspot more responsive and also adding some SEO features - Adding designer Drag & Drop features users can easily customize the HubSpot theme..etc?
Oct 29, 20217:49 AM - edited Oct 29, 202112:07 PM
Participant
iOS 15 - let’s talk about it
SOLVE
Hello all,
On the performance dashboard, in each specific email campaign, it shows opens by email client report at the bottom. It would be great if we could click on the bars in the chart to see who the IOS contacts are. Exporting that so we can create a list would then allow us to send A/B emails and separate that group in future emails.
Huge assumption here, but as that report with the email clients already exists in the email performance dashboard, perhaps its not a mountain to add in the above features as the connections are already there?
The challenge with a lot of these suggesstions (while at face value, they're great) is that this is reporting on the first open of an email by the recipient. What I mean by this is say you are sitting on a zoom meeting and your passively scolling through your inbox on your iphone and you open an email about an upcoming webinar that looks intriguing. At that point, the open fires and is recorded. BUT, you didn't register for the webinar because doing that on your phone would be a major PIA. So you don't delete the email and come back to it a few hours later and you open that same email on your desktop / laptop to register for the event and the open fires and is recorded again. The second open is a better signal, however, since the first open happened on an iphone, in the above report, it's indicated as such. So if you start treating the individuals differently that happened to open your email on their mobile device first, you could be leaving real engagement on the table. All in all, people have multiple devices and trying to pinpoint iOS users doesn't do a lot of good. I just assume everyone is using iphones which allows us to make better decisions to stregthen our email program.
Oct 26, 20219:55 AM - edited Oct 26, 20219:56 AM
Participant
iOS 15 - let’s talk about it
SOLVE
So in the Marketing Email Performance View where it shows "Opens by email client"...in our case as an example 34% Apple Mail....this does NOT include those that are filtered out by the bots setting, correct?
Pertaining to the iOS changes, I am looking at reporting and noticed some differences in terms between the email reporting and custom reporting. In the sent email page in the bottom chart called "Opens by Email Client" there is Apple Mail and Apple Mail (iOS) listed as browser types. Are both of these the browser types affected by the iOS change?
When I pull a list of email openers from Custom Reporting I get the browser names Apple Mail and Safari Mobile. Is Safari Mobile the same as Apple Mail (iOS)?
Hi Tom - Can you tell me if HubSpot can sync contact information for people who sign up for email marketing twice, using both their regular email address and also an address hidden by Apple's privacy protection. Will those get merged into one contact property?
How are "unengaged" contacts being affected? Since we cannot accurately continue to use open rates as a metric to track engagement, what do we do? Make our own list of unengaged users?