Apple’s Mail Privacy Protection: a concrete threat to the future of Email Marketing or a restriction that can be curbed?

Premise: One Mail Privacy Protection objective is to hinder the tracking of Apple Mail user behavior. This is why, as of now, it’s impossible to distinguish Apple users who have enabled Mail Privacy Protection from those who haven’t. This means that data reported below aren’t fully certain and accurate.

We’re constantly monitoring any Mail Privacy Protection impact, especially after major updates like the release of macOS Monterey. This has produced a “non-real” opening peak of 16% among MailUp customers and 41% among Apple users (regardless of email client and operating system). These percentages are certainly on the rise compared to those recorded after the first week from the release of iOS15.

Our deliverability consultancy actively supports a MailUp customer whose database has one of the highest shares of Apple users. In studying its Opening Trends, we found an increase in both the rates of unique openings on Apple domains (graph 1) and the rate of unique openings on all mailbox providers (graph 2).

The engagement trends (graph 3) also show a slight increase in the clusters of active and very active users compared to inactive ones, while the customer didn’t make significant changes to the target audience.

Overall, however, the number of MailUp customers impacted by MPP is very small: only about 30 customers recorded more than 20% of “non-real” opening rates. Further, just a few of these 30 customers are showing significant impact from their statistics.

We monitored the possible impact of Apple’s Mail Privacy Protection by conducting various analyses to grasp the extent of this change on data tracking activities. Daily and overall sending volumes from the MailUp customer base served as a reference for us.

Considering that Apple users represent approximately 25–30% of the MailUp daily customer base, here are the salient data we’ve identified:

• out of total MailUp customers, the rate of “not real” openings, i.e., presumably conditioned by Mail Privacy Protection, is around 6%
• out of total Apple users (regardless of the email client and operating system adopted), the rate rises to around 25%

To better understand the impact of the new Apple restrictions, we studied the performance of one of the MailUp customers with the highest rate of Apple users in the database.

Statistics of openings and engagement didn’t show any really alarming data. If we look primarily at the engagement trend, the impact of Mail Privacy Protection is practically irrelevant:

During last month’s Worldwide Developer Conference, Apple announced a series of new privacy features of the iOS 15 operating system update.

This looks normal and quite irrelevant for the Email Marketing world. However, Apple’s Mail Privacy Protection aims at limiting the amount of user data that can be collected through its email service.

The announcement immediately triggered a heated debate on how this may affect the future of email tracking activities and data-driven strategies.

Let’s take a step back: this isn’t the first time someone has tried to stop open tracking. Microsoft was, in fact, the first to introduce this concept with its Outlook clients. It disabled images by default so that users have to click to view it. What then seemed like the “doomsday of openings” has led marketers to create more sophisticated messages and recipients to disable the default image block.

Then it was Google and Yahoo!’s turn with image caching. The images were downloaded from the provider’s proxy rather than the final recipient. This triggered some fictitious opening every now and then. However, these providers didn’t mean to prevent the sender from accessing information about the opening. They just wanted to “mask” it by using the IP and useragent of the provider, rather than those of the recipients.

Lastly, Hey.com has launched a mail service that guarantees the “neutralization” of the tracking pixels. On the other hand, Sensorpro is becoming the vanguard of the Do Not Track movement. This proposes that the user consents to the monitoring through pixels right from the registration form.

Obviously, Apple’s initiative is part of a broader attempt, shared by various providers, to offer users greater control over their privacy. This is fundamentally correct and absolutely in line with the current evolution of privacy awareness.

These “privacy-first” features grow and appear differently. Granularity and freedom of choice left to end users make the difference, and these are two elements that shouldn’t be underestimated.

More, let’s not forget that several email service providers not only make their customers promptly informed on how to track data but also provide tools (such as preference centers). These allow for disabling this tracking for their addresses.

Apple’s Mail Privacy Protection makes it impossible to determine user behavior. It provides a background noise via a series of “not real” openings that are generated by Apple’s systems rather than users. This hinders the tracking back of a real action to a recipient.

Apple technology allows the latter to configure the client in two different ways.

• Case 1: They can view images of the emails received directly from their device (as before) or decide whether to go through an anonymous proxy integrated with the client.
  In the latter case, the IP and user-agent who read the email aren’t the real ones of the user. This is effective in preventing the geolocation and identification of the actual client, as well as the tracking of user behavior without first-party information.
  The date and time of the opening are reliable, though.
• Case 2: They can configure the client so that it arbitrarily downloads the email content within a variable timeframe after reception and independently from the user reading it.
  In this case, the IP and user-agent won’t be the real ones, as above, but not even the reading date and time can be considered reliable. All subsequent “real” readings by the user won’t generate calls on the service provider’s servers and will, in fact, be invisible.

Since the anonymized information is identically provided to the service provider in both cases, we can’t distinguish the two of them. In fact, this then makes all these openings “unreliable.”

Mail Privacy Protection affects first and obviously one of the email marketing metrics most used by marketers, i.e. the open rate.

Everything and then some has been said about this on-mail KPI. Many experts believe it to be an inaccurate metric.

From a certain point of view, this is true. The open rate is the “low-hanging fruit” of the metrics: easy to recover and beautiful (because a 40% open rate conveys a greater impression of success than a 3% click rate). It’s also useful for evaluating A/B tests and creating segmentations (between active and inactive users or according to the reading time slots). However, in terms of “accuracy,” this metric is often unsatisfactory: several anti-virus/anti-spam systems can simulate user openings. Some recipients read emails without viewing the images, and there are systems that block the tracking pixels. Nothing new, here.

Furthermore, this metric has created a trend. People overshadow the relevance of the message in favor of testing the subject lines with the best opening performances and privilege strategies to optimize their effectiveness. The goal here is just a quantitative increase in the open rate.

It’s also true that the open rate can be used with a grain of salt. It may represent the funnel starting point to get to the metrics that really matter: clicks, conversion, ROI, or the key to understand, free of background noise, if a certain change can help to hit the target.

1. Damaging the sender’s reputation and the ability to deliver messages

In any case, eliminating a “data point” makes marketers’ life tougher. They need as much information as possible – no matter how exact – to define their strategy.

On the one hand, in a sense we should “thank” Apple because its move will push marketers in considering more interesting and ROI-related metrics. On the other hand, though, most likely the number of emails will increase. This will clash against some engagement rate rules that typically define the sending frequency.

This could impact the sender’s reputation and the delivery of messages in the mid, if not immediate term. It’ll therefore be essential – for marketers who use openings and engagement metrics to determine sending frequency – to appropriately discuss all the variables with their email service providers and find the best strategy to protect their reputation in the subsequent months.

MailUp’s Deliverability Suite offers you a tailor-made consulting service and access to a wide range of advanced configurations that protect mailings and reputation. Discover all the advantages of this service!

2. Loss of information that the algorithms need

Now, openings are also meaningful for the most diligent Email Service Providers like MailUp. In fact, they allow for intercepting very low opening rate customers and verifying that they aren’t violating the terms of use. While not the most important signal, removing it would mean losing important information and having to re-calibrate the algorithms accordingly.

3. Drop of database cleanliness and quality

In my opinion, there’s one thing that’s even more problematic: this change will impact the engagement algorithms and the distinction between active and inactive recipients.

The main mailbox providers, including Apple, have always advised to keep your contact list “clean.” You need to unsubscribe the inactive ones to interrupt sendings that could deteriorate your reputation in the long run.

We also suggest contact unsubscription as a physiological and healthy element among the best practices of database building. Over the years, in fact, the MailUp customer base has showed us several concrete cases of customers who have improved their reputation and deliverability. They’ve used specific strategies for the re-engagement of inactive contacts and kept the database clean by facilitating unsubscriptions for uninterested users.

Indeed, openings have always served the aforementioned purposes. If, in the future, they can no longer be used to obtain these advantages, then we’ll need to find the way to replace them. Perhaps the mailbox providers will offer these metrics for a fee, like the Verizon Media Group has started to do. However, this solution, in addition to causing a potential conflict of interest, would raise an ROI problem, sparking discontent among many marketers who may not be willing to pay for this information.

But in the end, it’s fair to say, the final recipients never cared about the tracking pixel, and those who didn’t want to be tracked have already managed accordingly. Recipients care about a message content to respond to a need. Marketers – at least those who aren’t already doing so – should take a cue from this to focus on the relevance of the message and measure the metrics that really matter in terms of business growth and revenue. In the end, no one makes money out of the open rate.

Furthermore, Apple’s move can even help in breaking down the silos that still separate on-mail and off-mail metrics (page views, conversions, purchases, and revenue): today, more than ever, we need to improve these two worlds’ synchronization and ensure their full integration.

While waiting to grasp what’s really going to happen in September, let’s give you some advice to reflect and evaluate your strategies:

• start mapping inactive users and consider launching re-engagement campaigns: get them active again or unsubscribe for good.
• take other indicators into account more than openings:
  If the MailUp platform doesn’t track these indexes, then start evaluating how to integrate them into the platform.
• A/B tests may no longer work in determining the effectiveness of a subject line. Refer to new indicators to understand the A/B test impacts right away.
• If you’re using openings as a main marker of the active recipients of your list for tuning your sending frequency or cleaning up your list, then you may end up with more recipients than expected. Experiment how to combine this practice with the use of winback emails to successfully identify active users.

At MailUp, we’re already taking steps to timely adapt the platform tools to the new challenges of the email ecosystem.

Indeed, one of the benefits of a professional sending platform is that it filters the inconveniences of change for the customers while wisely and strategically navigating them through any transformation and unexpected future.