Why open rate is the wrong number to optimise for
Open rate has been mostly noise since Apple shipped Mail Privacy Protection in 2021, which pre-fetches every tracking pixel on Apple Mail regardless of whether a human opened the email. For a decade it was the metric every marketer reported, easy to measure, easy to compare, and it was deeply, structurally misleading even before Apple made it worse.
This post is about why, and what to track instead.
What open rate actually measures
When an email reports an "open," what's really happened is that the email client downloaded a small tracking pixel from your ESP. Historically that was a reasonable proxy for "a human looked at this email." A few things broke that assumption:
- Image blocking. Outlook (still 10–20% of B2B inboxes) blocks images by default. Those opens never get counted.
- Email apps that preload. Gmail's tabs feature, mobile apps that pre-fetch, corporate proxies, all of these can fire opens with no human involved.
- Apple Mail Privacy Protection (MPP). Since iOS 15, Apple Mail pre-fetches every tracking pixel for users who opt in to MPP. That's a lot of users. Your open rate now includes a fictional 100% open rate for that segment.
The net effect: open rate is an inconsistent blend of "people who actually looked" and "robots that pre-fetched." It moves for reasons that have nothing to do with your campaign quality.
What to track instead
Pick the metric that maps to the decision you're trying to make.
For campaigns: revenue per recipient (RPR)
RPR is total revenue attributed to a send, divided by the number of recipients. It punishes large unengaged sends and rewards tight, well-segmented ones. Use it as the primary scorecard for every broadcast.
For flows: conversion rate by step
For automated flows (welcome, abandonment, post-purchase), track conversion rate per step, not opens. You want to know which email in the sequence is doing the work, and which one is just absorbing list fatigue.
For deliverability: inbox placement + spam complaints
Use Google Postmaster Tools (free) and your ESP's spam complaint rate. Aim for under 0.1% complaints on broadcast sends. If you can afford it, a third-party inbox monitor (GlockApps, Litmus, etc.) shows where you're landing, Inbox vs Promotions vs Spam, by mailbox provider.
For list health: 30-day engagement
"Has this person clicked, ordered, or visited in the last 30 days?" If not, they shouldn't be in your broadcast segment. Sunset disengaged subscribers ruthlessly; deliverability rewards you for sending to people who want your email.
How to brief the team
Three lines, no more:
- We don't optimise opens because Apple breaks the metric.
- The new scorecard is revenue per recipient, by send.
- Anything we do that increases RPR is a win, even if open rate drops.
Then make sure your monthly report leads with RPR, not opens. What you measure is what gets optimised. Stop measuring the wrong thing.
Not sure where your email program actually stands? Run the free audit, ten questions on your automations, deliverability, and reporting, with a score out of 75 and a prioritised list of what to fix first. Or if you'd like a hand reviewing your setup properly, take a look at how our audits work.
Frequently asked
Not completely, it still has some signal for warmup campaigns and for catching catastrophic drops. But for week-to-week campaign comparison, Apple's Mail Privacy Protection makes it close to noise. Use it as a diagnostic, not a target.
Revenue per recipient (RPR) for campaigns, conversion rate by flow for automations, and inbox placement (via a deliverability monitor like Google Postmaster Tools) for sender reputation. These are what actually correlate with revenue.
Reframe it in dollars. 'A 40% open rate that drives $0 is worth less than a 20% open rate that drives $50 per send.' Most CEOs care about revenue, not rates, give them the metric that maps to the P&L.