How does email client caching affect metric accuracy?
Still have a question, spotted an error, or have a better explanation or a source we should cite?
Email open tracking works by embedding a tiny invisible image (a tracking pixel) in your message. When someone opens the email, their client loads that image, and your tracking system records an open. Straightforward in theory. In practice, caching breaks this.
When an email client caches images, it downloads and stores them on its own servers rather than loading them from your tracking pixel URL each time. The most significant example is Apple Mail Privacy Protection (MPP), which prefetches all email images through Apple's proxy servers, often before the subscriber even opens the message. The result: every Apple Mail user appears to have "opened" your email, whether they did or not.
Gmail also caches images through its own proxy (the Googlebot image caching system), but its behavior is different. Gmail caches images after the first open rather than before, so it doesn't inflate open counts in the same way MPP does. It does, however, mean a second open by the same person on a cached image might not register as a new event.
The practical impact: open rates for lists with high Apple Mail penetration can be inflated 20-40 percentage points compared to their real engagement. You can't fully trust open count as a measure of actual human engagement anymore.
Clicks are far more reliable. A click requires an actual human to tap or click a link. Caching doesn't affect click tracking in the same way (though bot clicks are a separate problem). If you want a signal of real engagement, weight click behavior over open rates.
For segmentation and re-engagement thresholds, use click data as your primary signal. Open data is still useful directionally, but don't suppress or re-engage based on opens alone.
Contributors
Who worked on this answer
Every name links to their profile. Every company links to their site. Real people, real accountability.
