With Apple Mail Privacy Protection, 90% of Apple Mail users will stop reporting email openings. What % of your database do they represent? Will working with the click rate be enough? After investigating the points of view of experts like Litmus, we reveal today the measures that Data Innovation will apply with our clients to continue having optimal performance in email marketing and CRM campaigns after the change.
What is the new Apple update about?
At its conference in June 2021, Apple announced several changes to its strategy to protect the privacy of its users. The new change will go into effect with iOS 15, iPad iOS 15, and macOS Monterey, likely in September 2021.
“In the Mail app, Mail Privacy Protection stops senders from using invisible pixels to collect information about the user. The new feature helps users prevent senders from knowing when they open an email, and masks their IP address so it can’t be linked to other online activity or used to determine their location.” Apple
With the new update, Apple will route emails through a proxy server, thus protecting the user’s IP address, to preload the content of the message, including tracking pixels, before delivering it to readers. This affects any email opened from the Apple Mail application on any device. Users will be able to choose between “Protect my Mail activity” or “Do not protect my Mail activity”, which means that this option will not be activated by default. But who is going to say “no”?
With Mail Privacy Protection it will no longer be possible to calculate users who open an email, the opening time, their location or type of device they use. Apple Mail’s share of the email client market share is 49% and according to the Litmus report, 90% of Apple Mail users will opt for the upgrade and email privacy protection. It is simply too high a percentage to ignore.
We have to point out that there are Apple users who do not use Apple Mail as an email application, but rather Gmail or Outlook in order to take advantage of calendar synchronization and integration with the other applications of the Google or Microsoft work suites. The question we ask ourselves is HOW MANY and how to detect them more accurately.
How will it affect open rate?
All emails sent to Apple Mail users who have chosen the new option will show as open, even if the recipient ignores it. That means that if your mailing list has a high percentage of Apple Mail users, the open rate can be inflated.
To roughly calculate the new open rate, Litmus offers a formula.
(Percentage of Apple Mail openers ÷ 100) x 0.9 x 75
+
Open rate today x (1 – [Percentage of Apple Mail opener / 100 x 0.9])
Where 0.9 means the expected adoption rate of the new update and 75 means the expected open rate of Apple Mail clients (these variables can be adjusted according to particular predictions).
At Data Innovation we analyze the data of individual users to decide their level of interest and activity according to the RFMi model. It means that a simple way to be able to count the users affected by the change will be that from a certain date they will begin to report an open rate of 100%.
That does not mean that from one day to the next we will have to forget about the open rate, but rather that we have to rely more on the click rate. According to Forbes, “Apple is forcing marketers to adopt a higher standard, innovate their email strategies and switch to more valuable metrics like clicks and engagement to demonstrate campaign effectiveness.” It’s a good point of view and a good time to improve the base metrics you use for decision making. For example: Did you know that there is a segment of users that we see in the databases of many of our clients that exclusively open emails? We label them as Window Shoppers and we reduce the frequency of interaction with them because in reality users that you cannot convince to reach the page you want to show them, means that they are not at all interested or do not have the means to be able to become a client.
In short – the effects on the part of your DB that uses Apple Mail, after Mail Privacy Protection
Checklist / action plan
The important thing to keep in mind is that our subscribers stay happy and continue to receive adequate and personalized content that they like so much and which they demonstrate by interacting with it (clicks, downloads, sales, etc.). One of the effects of the new update may be that users receive even more spam emails, as marketers will have one less engagement signal to follow.
In order to prepare for the new changes it is advisable to update and change certain techniques.
At Data Innovation we are dedicated to helping Marketing & CRM teams increase traffic, sales and new activations and optimize their KPIs. As a data, CRM and optimization consultant, we work with companies that send more than 10MM emails, agencies, eCommerce or retailers, and ESPs. And as partners, we rely on Tableau, Sparkpost, Netcore, Litmus, although we adapt to the technological environment of each client. The objective of our service is for your Marketing & CRM team to have visibility and perspective, apply expert knowledge, learn and put into practice pragmatic and innovative growth tactics, integrate the most advanced tools with a focus on the final results: more sales, more active users, more ROI.
Sources
The concept of data literacy is the ability to read, understand and make decisions based on data. Thanks to this ability, employees of any level can search for the right information in data, make decisions and convey the meaning of the information to others.
Some of the challenges that data literacy implies for companies are:
According to a recent study (by Censuswide) in which more than 7,000 managers participated, only 24% said they felt they had a good level of data literacy, despite 92% said that it is important that employees have data literacy, Only 17% say that their organization is taking significant steps so that staff can use data with greater confidence.
Darell Huff, an American writer and the author of the bestseller “How to Lie with Statistics”, the best-selling statistics book of the second half of the 20th century, describes some of the errors in the interpretation of statistics and in the easiness in altering the meaning of data.
In today’s article we bring you a summary of the best practices to understand data objectively according to the book “How to lie with statistics”:
When two variables X and Y are correlated, there are four possible explanations:
A. X causes Y
B. Y causes X
C. A third variable, Z, affects both X and Y
D. X and Y have no relationship
For example, when we hear that having regular physical activity habits is positively correlated with a longer life expectancy, we conclude that the more sport you do leads to a longer life expectancy. However, the results could be influenced by a third factor, such as diet or economic level. This third hidden variable can lead us to incorrect conclusions about causality.
It should be noted that in observational studies there are additional factors that we do not measure, therefore questions about causation can be answered by randomized controlled trials.
Humans like orderly and causal narratives, but the data doesn’t always say the same thing.
We cannot assume that correlations always follow the same positive or negative direction. Linear relationships are almost always only linear in a limited region of both variables. Beyond a certain point, the relationship can become logarithmic, disappear completely, or even reverse.
This can be observed in growth curves over time, in which, for example, there may be periods of linearity where growth occurs at a constant rate, but eventually, growth stabilizes because almost nothing continues to grow indefinitely.
As a first principle, the y-axis in a bar chart should always start at 0. Otherwise, it is easy to test an argument by manipulating the range, for example by converting minor increases to massive changes.
This is a technique widely used in the media and this happens because people do not read the information. Most people look at a graph and immediately draw a conclusion from the shape of the lines or bars, exactly as the person who made the graph wants.
When conducting a study, a sample is used, that is, a subset of the population intended to represent the entire population. This works well when the sample is large enough, but often due to limited funding or low response rates, psychological, behavioral, and medical studies are conducted with small samples, leading to questionable and unrepresentative results.
Humans are not very good at adjusting sample size when evaluating a study, which in practice means that we treat the results of a 1000-person trial in the same way as a 10-person trial. This is known as ‘sample size insensitivity‘ or ‘sample size neglect’.
The definition of average can vary greatly depending on the terms we use. The options that exist are the following:
Mean: add the values and divide by the number of observations
Median: order the values from least to greatest and find the middle
Mode: find the value that occurs most frequently
For example, the median and median median income in the United States differ by approximately $ 16,000.
It will be very important to know when “average” is specified, we must clarify if you are talking about the mean or the median because it can make a big difference.
The world is not symmetrically distributed and therefore we should not expect the mean and median of a distribution to be equal.
When looking at a statistic, the important question is often not what the value is, but how the current value compares to the previous value. In other words, what is the relative change compared to the absolute magnitude?
The data is often on scales that we are not familiar with and we need a comparison with other numbers to know if a statistic represents a real change. For example, is the 14,056,000 km2 area of the Arctic Ocean large?
Huff describes the idea of an “acceptable name” as one that is added to a study to give it an air of authority. Medical professionals (doctors), universities, scientific institutions and large companies have names that lead us to automatically trust the results they produce. However, many times these “experts” did not actually produce the work, but only tangentially participated and the name has been added to influence us.
Even when the results come from a confirmed “expert,” that doesn’t mean you should accept them without hesitation. The argument from authority is a fallacy that occurs when we assume that someone with greater power is more likely to be right. This is false because past success does not influence whether current results are correct.
Conclusions
In short, the book suggests that we keep a skeptical look towards any type of data. Any number represents a distillation of a set of data, which was taken from a sample of a population by error-prone humans, using imperfect tools, under constantly changing conditions at a single point in time.
All of this leads us to two conclusions:
If you put all your faith in a number, then you are over-tuned to the particular circumstances that produced the number.
Statistics and data are never purely objective. A statistic is an interpretation of uncertain data designed to persuade.
