Published 
Updated
Your open rate is higher than ever while your revenue stays flat. Since 2021, that gap has had a technical cause.
The open rate works with an invisible one-pixel image in your message. As soon as the email client loads that image, your software counts an open.
That measurement method worked reasonably well for years. Apple changed that in 2021 with Mail Privacy Protection, which preloads messages whether or not anyone looks at them. The image is then loaded while the recipient is still nowhere in sight.
By now, this covers a considerable share of the market. In May 2026, Apple Mail accounts for almost 65% of all registered opens, which means two out of every three opens in your report can come from preloaded messages.
As a result, the figure has mainly become inflated. Many businesses actually saw their open rate rise after 2021 and celebrated an improvement that was purely technical.
This page shows how to handle that figure and which numbers to steer on from here. The tactics that deserve attention stay. From now on, you simply judge them on something that holds up.
Apple fetches the content of a message via its own servers, before anyone opens it. All images are downloaded in advance, including the tracking pixel.
This causes three things at once. Your software registers an open that never actually happened. The time recorded belongs to the preload, so around the moment of delivery. And the IP address passed on belongs to Apple, which makes the location useless.
The feature is switched on for anyone who taps it during setup, which most people do. It also applies to everyone who uses the Apple Mail app, even when the email address is with Gmail or a custom domain. That last point often comes as a surprise.
Other companies followed with similar protection. The direction is clear, as privacy is gaining ground and tracking pixels are losing it.
It is important to know what still stands. Clicks come from a real action by a real person, so they remain reliable.
Fortunately, you can check this yourself. Four signals show how strongly it affects you.
Add those four together and you roughly know which part of your open rate is noise. Your own share of Apple Mail tells you how much weight it carries. Worldwide, Litmus measured almost 65% for May 2026.
Once you know who receives preloaded messages, you can set that group apart. That is the most important practical step you take.
Create two groups. The first consists of recipients whose opens are reliable. The second of recipients whose opens happen automatically.
For the first group, your open rate remains useful as a signal. For the second group, you let it go completely and steer only on clicks and purchases.
That distinction has practical consequences. Automated rules that mark someone as inactive based on opens backfire in the second group, because everyone there always opens. Inactive addresses then linger unnoticed in your list, which puts pressure on your deliverability.
So rebuild those rules around click behaviour and purchases. A recipient counts as active once they have clicked or bought within ninety days, whatever the open rate claims.
Many sending platforms now offer a ready-made filter for this group. With platforms that are still adding it, you build it yourself based on the detected email client.
The damage goes further than the open rate alone. Two other data points get muddied too.
The time of opening becomes useless for a large part of your list. Reports showing when your audience reads are skewed as a result. Optimising your send time based on opens steers you the wrong way.
The location disappears too. The IP address passed on belongs to Apple instead of the recipient, so country overviews based on opens say little.
So base your send time on clicks instead of opens. Click times come from real people and do show when your audience is active.
Get your location data from your sign-up form, your order data or your website statistics. Those sources are accurate.
Of all the figures, the unique click is the most reliable one left. A click happens because someone deliberately points at something.
Work with unique clicks instead of total clicks. Otherwise one enthusiastic reader who clicks seven times skews your picture. Unique clicks count people.
Put the ratio between clicks and opens next to it for the group where opens are reliable. That figure tells you how convincing your content is once someone has opened the message.
Also look at which links get clicked. A message with five links where one gets all the attention tells you exactly where the interest lies. That information feeds your next segmentation.
Keep an eye on one thing. Some security systems in business environments click links automatically to check them. That produces clicks within a few seconds of delivery, often on all links at once. Good sending software now filters this out.
A reply takes effort, time and willingness. That makes the reply rate the most honest signal email has. Yet it is rarely measured.
Its importance is growing too. Mailbox providers give replies a lot of weight when judging your reputation, because someone who replies demonstrably sees you as a wanted sender.
Increase your replies by asking for them. Ask one open question at the end of a message and send from an address a real person keeps an eye on. No-reply addresses shut off this signal completely.
Start with your welcome message. A question about what someone is working on right now gets replies exactly when willingness is at its highest.
Expect modest numbers. A reply rate of a few per cent is already strong for a newsletter. That small share gives you useful information plus a better reputation.
Three figures together tell you how fast your audience drops off. Unsubscribes, spam complaints and bounces. Add them up into one percentage and you have an early warning signal.
This sum works better than the three separate figures, because they point to the same underlying problem. Recipients who lose interest leave via whichever route is most convenient at that moment.
Opens are harder to measure, yet the subject line matters as much as ever. For real readers, it still decides whether the message gets read.
What works is clarity. Say what is in the message and what benefit it brings. Vagueness meant to spark curiosity brings opens in the short term and irritation in the long term.
Keep the length manageable. Around forty characters stays fully visible on almost any screen. Phones cut off longer lines.
Put the most important word first. The first two words decide whether someone keeps reading while scrolling through their inbox.
Personal works better than promotional. A line that sounds like a message from a person usually performs better than a line that sounds like an advert.
From now on, judge your test on clicks instead of opens. A subject line that gets plenty of opens and little return has set the wrong expectation.
Besides the subject line, the recipient sees two more things before deciding. Both are often overlooked.
The preheader is the text that runs on after the subject line in the inbox view. If it stays empty, email clients fill it with the first line of your message, which regularly produces something like "view this email in your browser".
Use that space to complement the subject line. The subject line opens the topic and the preheader adds the concrete benefit. Together they form one message.
The sender name carries even more weight than both. People first scan who something is from and only then what it is about. A recognisable name that stays the same builds recognition.
So choose consistency. A personal name followed by your company name works well and keeps recognition intact when a colleague sends now and then.
All the tactics combined come second to the question of whether your message fits the recipient. Relevance moves the figures far more than any subject line.
Three changes deliver the most. Send to smaller groups that have something in common. Connect with what someone did recently. And adjust your frequency to engagement, so active recipients hear more and quiet recipients are approached more gently.
An example makes it concrete. For years, an installation company sent one monthly message to everyone. After splitting into three groups, namely customers with a maintenance contract, customers considering one and people who were still exploring, each group got its own angle. The same effort, considerably more clicks.
That gain comes from fit rather than smarter wording. Segmenting your list usually gains you more than spending a week tinkering with subject lines.
So build on what you know about people instead. Preferences, behaviour and purchase history together form the material that creates relevance.
Replace your current overview with a setup that follows reality.
An average across your whole list hides exactly the information you want to steer on. Two segments moving in opposite directions together produce a flat line.
So split on at least three axes. On engagement, meaning active versus dormant recipients. On relationship, meaning customers versus people who are still exploring. And on origin, meaning per sign-up source.
That third axis often surprises people most. Sign-ups via a product selector usually behave completely differently from sign-ups via a giveaway. Only a breakdown per source reveals that difference.
Also look per message type. Transactional email, relationship email and offers should each have their own benchmark. Comparing them with each other creates apparent problems that are actually healthy.
Keep the breakdown manageable. Three to five cuts you review monthly steer you better than fifteen you set up once and then leave.
Many subject line tests still judge the winner on opens. Since 2021, that has produced skewed results.
So judge the winner on unique clicks. A line that gets plenty of opens and little return has raised an expectation that undermines the message.
Make sure you have enough volume. With a test group under a thousand recipients per variant, the differences are rarely reliable, so you end up recording coincidences as lessons.
Test one thing at a time. Two variants that differ in length, tone and offer produce a winner whose cause remains unclear.
Repeat the winning direction a few times before you make it your standard approach. One result is an observation, three in a row make a pattern.
Record every result in a simple overview. After a year, you own knowledge that is specifically about your audience.
Overviews with average open rates per industry are everywhere. Since preloading messages became the norm, those numbers say remarkably little.
The problem lies in the composition. Every list contains its own share of Apple Mail, so every average contains its own amount of noise. Two businesses with identical performance therefore report very different figures.
Use such overviews as a rough direction at most. When your click rate is far below a widely supported average, that is a signal to take a look. That is about where their value ends.
Compare with yourself instead. Your own development over twelve months, per segment, tells you everything an industry average tries to suggest.
Also build your own benchmarks. Record per message type what a normal result is for you and measure against that. That is the only yardstick that fully fits your audience.
The click is the beginning of the story and rarely the end. What happens on your site afterwards tells you whether the message did its job.
So add recognisable tags to every link. Your analytics package then shows per message and per flow how many visitors came in, how long they stayed and which step they took.
Then look beyond the first page. A message that delivers plenty of clicks along with visitors who leave straight away points to a promise the landing page has yet to keep.
Link it to a concrete action. A quote request, a completed form or an order shows which part of your traffic actually delivered something.
Compare that per segment. Two groups with a similar click rate can differ sharply in what happens next. That second step is what determines your result.
Keep these figures per month. A year from now, you will have a picture of which message type contributes structurally, separate from the noise in your open rate.
More in this series on email marketing
The next step
Discover your biggest growth opportunities with a free opportunity analysis (worth €416) — a tailor-made report, free of charge and with no obligation.
About the author
Dominic Pietersz
Your dedicated point of contact
Dominic helps SMEs grow through Google and other channels. With his international experience, he turns challenges into concrete growth.
Follow Dominic on LinkedInInspiration · When our own campaigns had open rates well above 50%, I wanted to share the secret. That is how this piece about raising your open rate by 20% came about.
