Does MPP block all tracking?

No. Apple's Mail Privacy Protection (MPP) breaks open tracking and hides recipient IP addresses, but it does not block clicks, replies, unsubscribes, or anything that happens on your website after someone leaves the inbox. You still have plenty of signal. You just need to stop treating open rate as truth.

Here is what MPP actually does. When an Apple Mail user has it turned on, Apple pre-fetches all images in the email through a proxy server, usually within minutes of delivery, whether the recipient ever opened the message or not. That means your tracking pixel fires for almost everyone. Open rates inflate. Timestamps become useless. The IP you see belongs to Apple's proxy, not the recipient, so you cannot infer device, location, or time zone from it. Apple announced this at WWDC 2021 and it has been on by default in the setup flow ever since iOS 15. See Apple's own description of MPP for the official wording.

The share of mail this touches is large. Litmus has tracked Apple Mail at roughly half of all opens for years, and most of those users accepted the MPP prompt. So if your audience skews consumer or mobile, assume the majority of your "opens" are bot opens from Cupertino.

Here is what MPP does not touch:

• Clicks. When someone taps a link, their browser still hits your redirect, so click tracking, UTM tags, and link-level attribution all work normally.
• Replies. Direct human signal. No proxy in the middle.
• Unsubscribes. The one-click unsubscribe header and your hosted unsubscribe page both still fire.
• Website behavior. Sessions, add-to-carts, purchases, account logins. None of that runs through Apple.
• Conversion events. Whatever you measure on your side post-click is unaffected.

So the practical change is in how you segment and report, not in what you can measure. Stop using "opened in last 30 days" as your engagement gate for Apple domains. It is meaningless for them. Use clicks, site visits, or purchase recency instead. If you must keep an open-based segment for non-Apple mailbox providers, split it by domain. Gmail, Yahoo, Outlook opens are still mostly real (with their own caveats around image caching and prefetch, see Gmail's image proxy docs).

A few things people get wrong about this:

1. "MPP killed deliverability data." It did not. Spam complaint rate, bounce rate, and unsubscribe rate are all unaffected. Those are the metrics Gmail and Yahoo actually use to decide where you land, not opens. See our take on whether Sender Score is a Gmail metric for what the providers really watch.
2. "I should just remove the tracking pixel then." No. The pixel does not hurt you, and for non-MPP recipients it is still useful. We unpack that in Myth: removing the tracking pixel improves inbox placement.
3. "Resending to non-openers is still a clean play." Risky now. Half your "non-openers" are actually engaged Apple users whose opens never registered, and the other half might be truly disengaged. Read does resending to non-openers help or hurt before you set that up.
4. "Inactive subscribers based on opens are safe to keep mailing." Also risky. If you cannot tell who is really inactive, you are flying blind on list hygiene. See do inactive subscribers hurt deliverability.

The short version: MPP did not blind you. It killed one metric that was already noisy and forced you to look at the ones that actually correlate with revenue. Clicks, replies, site behavior, complaints. Those still tell the truth.