If you check your Substack or email stats and see a 40-45% open rate, you might feel pretty good about that number. Here's the uncomfortable part: a big chunk of it isn't real. Not fraud, not a bug — just a privacy feature quietly breaking the metric almost every creator still treats as their main scoreboard.
Back in September 2021, Apple shipped Mail Privacy Protection (MPP) with iOS 15. Before that, "open rate" worked the way you'd assume: a tiny invisible tracking pixel loads when a human actually opens the email, your email provider logs it as an open. Simple, and reasonably accurate.
MPP broke that quietly. When a subscriber has it turned on (and it's on by default for a lot of Apple Mail users), Apple doesn't deliver your email straight to their inbox — it routes it through Apple's own proxy servers first. Those servers preload everything in the email, including your tracking pixel, whether or not the person ever looks at it. So your dashboard logs an "open" that's actually just Apple's server saying hello, not a human reading anything.
The scale of the distortion is bigger than most people realize. Industry-wide, average reported open rates jumped from roughly 20-25% before 2021 to 35-45%+ industry-wide today — and that jump lines up almost exactly with iOS 15 adoption, not with newsletters suddenly getting better.
It's also not purely an Apple-only issue anymore, though the mechanism is different. Gmail has run its own image-caching proxy for over a decade (originally for security and IP-masking, not privacy timing), but it works differently: Gmail only logs a real "open" the first time an actual person triggers it, then serves cached images on repeat views without re-counting them. That's a smaller, more limited distortion than Apple's — Apple's MPP preloads the pixel at delivery time for every recipient with it enabled, whether or not anyone ever opens the email. Yahoo runs similar image proxying to Gmail's. So the inflation isn't identical across providers, but the practical result is the same: none of the major inboxes deliver a fully "clean" open signal anymore, and Apple Mail users are by far the biggest source of the distortion.
Substack writers lean on open rate harder than most other creators, because it's often the only performance number visible right on the dashboard without digging. That makes it tempting to treat "open rate went up" as "the newsletter is working" — but if your subscriber base skews toward Apple Mail users (and a lot of tech/newsletter-reading audiences do), a rising open rate might mean nothing more than "more of my list happens to use an iPhone."
This is also why comparing your open rate to some generic "good open rate" benchmark you found on Google is close to useless. Benchmarks published before 2021 and after 2021 aren't measuring the same thing anymore, even though they use the same label.
None of this means your numbers are meaningless — it means the wrong number has been getting the spotlight. A few that hold up better:
Open rate isn't worthless — it's still a rough pulse check that your emails aren't landing in spam. But it stopped being a reliable engagement metric years ago, and most "here's a good open rate" advice floating around still hasn't caught up. If you're logging your weekly numbers anywhere, it's worth tracking CTR and replies alongside open rate, not instead of it — and judging your own progress against your own past weeks, not against a benchmark that may have been inflated since before you even started.
Sources: Beehiiv's Apple MPP breakdown, Paubox's explainer on how Apple Mail Privacy Protection works, and Meettie's 2026 email benchmark report on inflated open rates post-iOS 15.
Turn your own content into a week of platform-ready posts.
Try the Repurposer free →