Email Open Rates Guide: What They Mean After Apple Privacy

Open rate used to be the slide everyone put in a QBR. Then Apple started prefetching mail, and a chunk of “opens” stopped meaning a human saw your subject line.

This email open rates guide is for operators who still get asked “what is our open rate?” and need an honest answer: use it as a smoke alarm, not a steering wheel.

If opens collapsed and clicks did not, you may have a deliverability problem. If both are fine and revenue is not, you have a merchandising problem. Those are different jobs.

What an “open” is now

An open is a tracking pixel loading. That can mean:

• A person opened the email.
• Apple Mail Privacy Protection fetched the pixel in the background.
• A security scanner opened it.
• Images loaded in a preview pane.

So a 45% open rate can be “the list is healthy” or “a lot of iPhones prefetch.” You cannot tell from the number alone.

Clicks still need a human. Placed orders still need a human with a cart. Start there.

Industry “average open rate” tables are still useful for order of magnitude (if you are at 8% something is on fire). They are not a target to bonus people on.

When open rate still matters

Treat it like temperature. You do not manage a factory by staring at a thermometer all day. You glance when something smells wrong.

Sudden drop across all sends. Not one campaign. All of them. That is inbox or domain, not a subject line. Check authentication (SPF, DKIM, DMARC), list hygiene, and whether you blasted unengaged. That is email deliverability work.

One campaign tanks vs last week’s similar send. Maybe creative. Maybe you sent at 5 a.m. Maybe the segment was wrong. Look at clicks before you rewrite the brand voice.

iOS vs Android split if you have it. A huge iOS open rate with dead clicks is prefetch. Do not “optimize subject lines for Apple.”

Welcome vs campaign. Welcome should beat campaigns. If it does not, the first impression is broken — see automation flows.

Metrics that should be on the weekly slide

Steal this table for the meeting that currently only shows opens.

Weekly email scoreboard
- Flow revenue (welcome, cart, post-purchase, winback)
- Campaign revenue
- Revenue per recipient (campaigns)
- Placed order rate (flows)
- Spam complaints
- Unengaged % of the list

If you cannot produce that from Klaviyo in fifteen minutes, the account is not instrumented. That is a Klaviyo setup issue, not a “we need more content.”

Holdouts. The adult version of analytics. 10% of a segment does not get the campaign. Compare orders. If you cannot stomach a holdout, you do not actually want measurement. You want a story.

Cohorts. Repeat purchase 30/60/90 by first-touch flow. Opens will not tell you if post-purchase is creating a second order. Retention strategy will.

How to run subject tests without worshipping opens

You can still test subjects. Change the rules:

• Primary metric: clicks or orders, not opens.
• Same send time, same segment, enough volume. A 2-point open win on 900 sends is noise.
• Do not test punctuation as a personality. Test a specific promise vs a vague one.

Example:

• A: “New arrivals”
• B: “The navy coat is back in 38R”

B might lose opens and win revenue. That is a good test. That is the whole point of A/B testing email campaigns.

A practical 30-day cleanup

Week 1. Screenshot the last 30 days: opens, clicks, orders, complaints. If orders are missing from the slide, add them before you debate subjects.

Week 2. Suppress or sunset unengaged (90–120 days). Watch complaints and delivery, not the vanity open increase from mailing fewer ghosts.

Week 3. Fix welcome and cart tracking. If flow revenue is a black box, nothing else in analytics is real.

Week 4. One holdout on a campaign. Live with the number.

Opens are not useless. They are just no longer the product. If you want an account scored on revenue leaks instead of a benchmark PDF, book the audit.