Apple Mail Privacy Protection arrived with iOS 15 and macOS Monterey in 2021. When a user turns it on, Apple’s servers download the remote content of incoming emails, including the tiny tracking images that record opens, through a proxy. That happens whether or not the person ever reads the message, and it hides their IP address, so senders also lose reliable location and device data for those opens.
It applies to the Apple Mail app, not to an email provider. Someone using a Gmail or Rediffmail address inside Apple Mail on an iPhone is affected; someone reading the same Gmail address in the Gmail app is not. So the impact depends on how many of your subscribers use iPhones and Macs with Apple Mail. For many Indian lists, which lean towards Android and the Gmail app, the effect is smaller than for lists in some other countries, but it is rarely zero.
The practical consequence is that open rate now overstates real reading. Anything that relies on an open is affected: “opened in the last 90 days” segments, re-engagement rules, open-based A/B test winners, send-time predictions and automation branches like “if opened, send X”. An inactive subscriber may look active because their phone fetched every email automatically.
The fix is to lean on stronger signals. Clicks, replies, orders, form submissions and site visits can’t be faked by a proxy. Use opens to watch trends over time rather than as an exact figure, decide content tests on clicks, and when you build an inactivity segment, combine “no opens” with “no clicks” and “no orders” so you don’t keep mailing people who only appear to be engaged.
A Bengaluru software company sees its open rate jump from 22% to 41% within a month without changing anything. Click rate stays flat at 2.5%. The rise comes from Apple Mail users with privacy protection on, so the team switches its A/B tests to pick winners by clicks.
In MailNesto you can pick A/B test winners by clicks instead of opens, and campaign reports show click rate and click-to-open rate beside opens so you can judge results on more than one signal.