How does Apple’s MPP confuse engagement-based filters?

Imagine you have a subscriber at captain@deepcurrent.io. They use Apple Mail on their iPhone. Your ESP reports that they opened your last twelve campaigns. Sounds great. But did they actually read any of them?

That's the core problem with Apple Mail Privacy Protection (MPP), which launched in 2021. When a user has MPP enabled, Apple's servers pre-fetch your email content, including the tiny tracking pixel that fires an "open" event, before the person even touches the message. So your ESP logs an open. The subscriber may have never actually seen your email.

Now multiply that across every Apple Mail user on your list. Some research has put Apple Mail's market share at 50% or more of all email opens, which means a huge chunk of your reported opens were never real human reads to begin with.

Here's how that confuses engagement-based filters. Filters like those used by Gmail have historically used open behavior as one signal among many to judge whether your emails are wanted. But those filters are looking at patterns across many users. If MPP inflates open signals for a large portion of your list, the filter's model of "what an engaged sender looks like" gets distorted data fed into it. Filters that leaned heavily on opens had to either adjust how much weight they gave to open signals or risk rewarding senders whose apparent engagement was partly a machine artifact, not human interest.

The practical consequence is that engagement-based filtering shifted toward signals MPP can't fake. Clicks matter more now because MPP doesn't click links. Replies, forwards, and moves to inbox all carry more weight because they require a real human action. Spam reports and deletions without reading still register cleanly too, which is a good reason not to ignore the negative signals.

For senders, the practical fix is to stop treating open rate as a proxy for subscriber health. A subscriber you haven't heard a click from in six months is likely disengaged, even if your ESP shows them opening regularly. Segment by click activity and real engagement signals instead. Your list hygiene strategy should follow click behavior, not open rates.

If you're not sure how much of your list is Apple Mail users, your ESP's domain-level reporting can usually break that out. And if your open rates look suspiciously high compared to your click rates, MPP is probably the culprit.