Apple Mail Privacy Protection and open rates

Opened at 3:12 am.
Read by nobody.

Apple Mail Privacy Protection downloads the images in an email, including the invisible pixel that counts an open, on Apple’s servers, whether or not anyone reads it. Since iOS 15 in September 2021, everyone who turns it on shows up as an open on every email, from an Apple address that hides where they are, and their real opens never reach you. Apple accounted for 62% of the opens Litmus tracked in July 2026, so your open rate can be mostly machines. The calculator below estimates the human open rate behind yours. After it, you’ll find which numbers still mean something (clicks, replies and sales), and how Builderify counts.

By the Builderify team

Illustration · One email to one Apple Mail reader with protection on.
01 / How it works

How Apple Mail Privacy Protection
changes what an “open” means.

An open has always been a guess: the moment an email’s images load. Apple announced in June 2021 that Mail Privacy Protection “stops senders from using invisible pixels to collect information about the user”. It does that by loading the images itself.

  1. 01

    Delivered

    Your email reaches someone who reads their mail in Apple Mail, with Mail Privacy Protection on.

  2. 02

    Apple fetches it

    Apple’s servers download the email’s images, your tracking pixel included, whether or not anyone opens it.

  3. 03

    You see “opened”

    Your email tool logs an open from an Apple address. Apple hides the reader’s IP address, so the location and device are Apple’s too.

  4. 04

    They read it, or don’t

    Either way, nothing more reaches you. Repeat opens in Apple Mail aren’t returned to the sender.

62%

of the opens Litmus tracked in July 2026 were in Apple Mail, protected opens included.

55–60%

of all email opens are affected by Mail Privacy Protection, by Litmus’s estimate.

Any account

read in Apple Mail, Gmail addresses included. Opens in other apps, such as Gmail on the web, are counted the usual way.

Since 2021

when it arrived with iOS 15, iPadOS 15 and macOS Monterey. Every open recorded before then is a different measure.

02 / Open-rate calculator

What’s your real open rate
after Apple’s opens?

Enter one campaign’s numbers from your email tool, and either the share of your list that reads in Apple Mail or, if your tool labels them, its count of Apple or machine opens.

Apple’s share

Your tool’s email client report shows Apple’s share of your opens. Across Litmus’s data, protection touches 55–60% of opens; a list of mostly Gmail or company addresses will be lower.

Reported open rate
62.0%

80.6% of these opens are Apple fetching the email.

Estimated human open rate
24.0%

Among readers whose opens can still be seen.

Opened by Apple Opened by a person Not opened, or not seen 1 square = 1% of delivered
Click-through rate
2.9%
Clicks ÷ delivered
Click-to-open, reported
4.7%
Dragged down by Apple
Click-to-open, human
12.1%
Estimated

How it’s worked out: Apple’s opens (delivered × Apple share, or your tool’s count) come off the top, and the human rate is what’s left of your opens divided by what’s left of your list. It assumes everyone using protection registers an open, and that they read about as often as everyone else. It’s an estimate to compare campaigns with, not a measurement.

03 / What still works

What to measure instead of open rate,
number by number.

Opens still exist; they just stopped meaning “read”. The safest numbers are the ones a person has to do with their hands. If an open sets anything off in your tool, that’s where to look first.

  • Open rate Broken Every Apple Mail reader with protection on counts as an open on every email. Compare a send only with earlier sends to the same list, and expect it to drift as more readers switch to Apple Mail.
  • Subject-line tests judged on opens Broken A large share of the “opens” never saw the subject line at all. Pick the winner by clicks, or by sales.
  • Open time and location Broken The time is when Apple fetched the email, and the address is Apple’s. Use clicks for timing, and the signup form or order for location.
  • Resend to non-openers Bent Apple readers all look like openers, so they never get the resend. Mailchimp says so of its own resend feature. Resend to non-clickers, or not at all.
  • Automations triggered by an open Bent Mailchimp warns they go to a larger audience than intended. Trigger on a click, a signup or a purchase.
  • Removing “inactive” subscribers by opens Bent Apple readers never look inactive, however long they ignore you. Judge activity by clicks, replies or purchases, over a longer period.
  • Click-through rate Still works People click; Apple doesn’t. Some company email filters do follow links before delivery, though. Discount clicks that land within seconds of sending, or on every link at once.
  • Replies, unsubscribes, complaints and sales Still works A person has to do each of these. Your best signals, and the ones mailbox providers watch too.
  • Bounces Still works They come from the receiving server, not the reader. Read the code, and act on it.
Clicks lie a little too

Some company email filters, Microsoft’s among them, follow the links in a message to check them before the person sees it. Those visits arrive within seconds of sending, often on every link at once, and from data centres rather than homes. That pattern is how to discount them. The click-quality guide shows the same signals for paid traffic.

Deliverability isn’t an open rate

A drop in opens used to hint at the spam folder; with Apple in the mix, it rarely moves. Watch the things that do: bounce codes, spam complaints, and whether SPF, DKIM and DMARC still pass on a real message.

04 / With Builderify

Our open rate is inflated too.
Here’s what we built around it.

Campaigns and Sequences count an open the way most tools do, with a tracking image. Apple’s preloads land in that count, and we don’t pretend otherwise. What we can do is make sure an open never decides anything.

For the number that matters, put the links you care about through a short link, one per email or one per offer, as the welcome sequence template suggests for its links.

How Builderify counts

  • Counts one open per person per email, the first time the tracking image loads; repeat opens are ignored
  • Nothing sends, stops or segments on an open, so Apple’s preloads can’t set anything off
  • The unsubscribe link opens a page with a button, so a filter following links can’t unsubscribe anyone; mail apps use the one-click header
  • Replies arrive in your Inbox, and bounces and unsubscribes take people off sequences
  • Short Links in an email count unique clicks, with proxy, VPN and data-centre flags shown alongside

Know this first

  • Apple’s preloads count as opens, and there’s no label to tell them apart
  • So the Open rate on your dashboards is inflated like everyone else’s
  • No click tracking inside campaign or sequence emails yet; use a short link
  • A short link counts clicks on the link, not which subscriber clicked
Questions

Open rates and Apple, before you ask.

Yes. When it is on, Apple's servers download every email's images, including the tracking pixel that counts an open, whether or not the person reads it. Those readers show up as opens on every email, so reported open rates go up, and their real opens are never reported.

Some tools mark them. Klaviyo labels opens as Apple Privacy Open true or false, SendGrid sets sg_machine_open, and Mailchimp can leave them out of reports for emails sent since 22 June 2024. None can tell whether a protected reader also read the email. Builderify doesn't label them.

Only those who read their Gmail account in Apple Mail with protection on. Opens in other apps, such as the Gmail app or Gmail on the web, are counted the usual way.

Things a person has to do: clicks, replies, purchases, unsubscribes and spam complaints. Judge subject-line tests by clicks, trigger automations on clicks or purchases rather than opens, and judge whether a subscriber is still active by clicks or replies over a longer period.

Take off the opens Apple produces: delivered emails times the share of your list using Apple Mail with protection on, or your tool's count of machine opens. Then divide the opens left by the emails left. With 2,000 delivered, 1,240 opens and half the list on Apple Mail, that is 240 of 1,000, or 24% rather than 62%.

Stop losing the visitors
you already paid for.

More leads from the traffic you already have, and follow-up that sends itself. Free to start. No card, no monthly fee.