Why open rates are unreliable
Apple opens your email before the recipient does.
Open rate is the number everyone quotes and the one you should trust least. It has been broadly unreliable since 2021, and the reason is mechanical rather than a matter of opinion.
How an open is measured
There’s no way to ask an email whether it was read. Instead every message carries a tiny invisible image, and loading it counts as an open.
That makes an open a proxy, and a leaky one in both directions:
- Undercounts anyone whose client blocks images — they can read the whole thing and never register.
- Overcounts anything that loads images automatically, whether or not a person looked.
What Apple changed
Apple Mail loads the image before the recipient opens anything
Since Mail Privacy Protection in 2021, Apple Mail pre-fetches images for messages it receives — so an open is recorded whether or not the person ever looks. Apple Mail is a very large share of email, so a large share of your “opens” are machines.
This is why open rates across the industry jumped at that point without anybody’s email getting better, and why comparing your rate with a benchmark published before 2021 is meaningless.
Use clicks instead
A click is a person deciding to do something. Nothing pre-fetches a click. It’s a smaller number and an honest one.
Which is why reports here lead with clicks, why engagement segments can key on clicks, and why an in-flow test defaults to clicks rather than opens.
Click-to-open rate — clicks as a share of opens — is still worth a glance for how compelling the content was once seen, but it inherits the same distortion in its denominator.
Opens aren't worthless
Two things they’re still good for:
- Comparing like with like. Two subject lines on the same audience in the same week share the same distortion, so a big gap still means something.
- Spotting a collapse. Opens falling off a cliff usually means delivery trouble, not disinterest.
What they can’t do is tell you how many people read your email.
FAQs
Should I stop looking at open rate?+
Look at it, don’t decide on it. Judge campaigns on clicks and, where you can measure it, what happened afterwards.
Does this affect device and location stats?+
Yes — which is why the Audience breakdowns are computed from clicks. Apple proxies opens through its own servers, so device and location taken from an open are partly fiction.
Can I turn open tracking off?+
Not currently. It still has the comparative uses above, and it’s what the industry expects to see on a report.