What Is Open Rate? Why the Email Metric Misleads and What to Use

Stop treating open rate as your main email metric. I know it’s the first number every email tool shows you, and the one most teams paste into monthly reports. But for many lists today, a rising open rate tells you more about which email app your subscribers use than about whether anyone read your email.

That’s not because open rate was always useless. For years it was a decent, if rough, signal. What changed is that privacy features, most notably Apple Mail Privacy Protection, now load tracking images automatically, so emails can count as opened when no human has looked at them. Subject line tests, “engaged subscriber” segments and send-time decisions built on opens are all quietly affected. This guide explains how open rate works, why it misleads, what it’s still good for, and what to measure instead.

What open rate actually means

In plain terms, open rate is the share of people who received your email and opened it.

Formally, email open rate is the number of unique opens divided by the number of emails delivered, expressed as a percentage. Unique opens count each recipient once, however many times they open the email. Total opens count every open, including repeats. Most email platforms report unique opens as the headline figure. Some divide by emails sent rather than delivered, which makes the rate slightly lower when some emails bounce, so check which definition your tool uses before you compare reports from two platforms.

The nuance matters more than the formula. An “open” isn’t really someone opening an email. It’s a tiny image in the email being downloaded from the sender’s server. When the image loads, the platform records an open. When it doesn’t, no open is recorded, even if the person read every word. So open rate has always been a measure of image loading, standing in for reading.

How open rate is tracked

Every tracked email contains a tracking pixel, a transparent image one pixel wide with a unique address for each recipient. When the email app loads that image, it requests it from your email platform’s server, and the platform logs an open for that recipient.

The formula is:

Open rate = Unique opens ÷ Delivered emails × 100

With example numbers, if 6,000 emails were delivered and 1,500 recipients opened, the open rate is 1,500 ÷ 6,000 × 100 = 25%.

This method has always had blind spots. If someone’s email app blocks images by default, a real read may not be counted. Plain-text emails have no image, so opens can’t be tracked at all. And anything that loads images automatically, without a person looking, gets counted as an open.

Why open rate misleads now

The biggest change came when Apple introduced Mail Privacy Protection for its Mail app. Apple Mail users are prompted to turn it on, and many do. When it’s on, Apple’s servers download the remote content in emails, including tracking pixels, in the background, often soon after delivery. The email platform sees the image load and records an open, whether or not the person ever reads the email. It also hides the person’s location and the real time they opened.

Other sources add more noise in both directions. Security systems at some companies scan incoming emails and load their images or links automatically, which can register opens and even clicks. Some email apps and workplace setups block images, which hides real reads. Gmail also loads images through its own proxy servers, though in that case an open is generally still recorded when the person views the email. Different lists have different mixes of these apps and systems, and the mix changes over time.

The result is a metric that’s inflated for some audiences, deflated for others, and inconsistent from month to month. It can no longer be compared reliably between lists, between brands or even across long periods for the same list.

What open rate is still useful for

Open rate isn’t worthless. Used carefully, it still does a few jobs:

• A deliverability alarm. A sudden, sharp drop across all email apps can mean your emails have started landing in spam. That’s worth investigating immediately.
• A trend within one list. Week to week, for the same audience, big movements can still hint at something changing.
• Segment comparisons with machine opens removed. Many email platforms now flag or separate opens likely caused by privacy features. Filtered opens are more meaningful than raw ones.

What it can’t do is prove that people read your email, tell you which subject line truly persuaded more people, or tell you who’s still interested.

A worked example: a bookshop newsletter

This is a hypothetical example with example numbers, for an independent bookshop in Chennai with a weekly newsletter sent to 8,000 subscribers.

For a long time the newsletter’s open rate hovered around 24%, with about 3% of recipients clicking. Over a few months, as more subscribers used Apple Mail with privacy protection on, the open rate climbed to 47%. The owner was delighted and told her team the new newsletter design was working.

Clicks told a different story. They stayed at around 3% of recipients, roughly 240 people a week. Sales from the newsletter didn’t move either. The extra opens were mostly machine loads.

Two decisions had gone wrong because of it. A subject line test picked version B as the winner because it had more opens, but version A had driven more clicks and more book orders. And her “inactive subscriber” clean-up, which removed people who hadn’t opened in six months, was now keeping thousands of disengaged Apple Mail addresses, because their emails were always marked as opened.

She changed three things. Subject line tests now choose winners by clicks. The inactive segment is now based on no clicks, visits or purchases in six months. And the monthly report leads with clicks, orders and revenue per email, with open rate at the bottom as a deliverability check.

Open rate vs click rate vs click-to-open rate

	Open rate	Click rate	Click-to-open rate (CTOR)
Formula	Unique opens ÷ delivered × 100	Unique clicks ÷ delivered × 100	Unique clicks ÷ unique opens × 100
What it suggests	The email was displayed	Someone took action	How persuasive the content was to openers
Affected by privacy features	Heavily inflated	Much less affected	Deflated, because inflated opens are the base
Best used for	Deliverability alarms, filtered trends	Judging emails and subject lines	Comparing content across emails to the same list

Click rate isn’t perfect either. Security scanners can generate false clicks, and some valuable emails, like a delivery update, don’t need clicks at all. But a click takes intent, so it’s far harder to fake by accident. For most marketing emails, clicks and what happens after them are the honest measures of whether the email worked.

How to measure email performance properly

1. Lead with clicks and conversions. Put unique clicks, orders or leads, and revenue per email at the top of every report. Use UTM tags so your analytics can tie email visits to results.
2. Filter machine opens where you can. If your email platform can separate privacy-driven opens, use the filtered figure for anything that matters.
3. Test subject lines on clicks. Set A/B tests to pick winners by clicks or conversions. Confirm any open-based winner also won on clicks.
4. Define engagement by action. Build “engaged” and “inactive” segments on clicks, site visits, purchases and replies, not opens.
5. Watch deliverability through several signals. Track bounces, spam complaints, unsubscribes and the sender tools the big inbox providers offer, alongside sudden changes in opens.
6. Compare only with yourself. Measure each email against your own recent sends to the same list, not against published averages.

Common open rate mistakes

Celebrating a rising open rate. If clicks and sales haven’t moved, the rise probably reflects privacy features, not better emails.

Choosing subject lines by opens. Machine opens add random noise, so the “winner” may be luck. Choose by clicks or conversions.

Cleaning your list by opens. Subscribers whose emails are always machine-opened never look inactive, so disengaged contacts stay on the list and can hurt deliverability. Use clicks and purchases instead.

Comparing with industry averages. Published open rates mix lists with very different email app mixes and definitions. They tell you almost nothing about your own performance.

Triggering automations on opens. A follow-up that goes to everyone who opened the first email will also reach people whose email app opened it for them. Base follow-ups on clicks or visits so the message fits what the person actually did.

Timing sends by open time. Privacy features can hide when people really opened, so “best time to send” reports based on opens may just show when Apple’s servers fetched images.

How to tell if your email programme is healthy

Instead of chasing an open rate number, watch these signals:

• Clicks, orders and revenue per email are stable or rising against your own history.
• Unsubscribes and spam complaints stay low after each send.
• Your engaged segment, based on clicks and purchases, is growing.
• There are no sudden, list-wide drops in opens that could signal spam-folder placement.
• People reply, forward and buy, which no privacy feature can fake.

Key takeaways

• Open rate is unique opens divided by delivered emails, but an “open” is really a tracking image loading.
• Apple Mail Privacy Protection and other automatic image loading inflate opens, so a higher open rate may not mean more reading.
• Use open rate as a deliverability alarm and a filtered trend, not as your measure of success.
• Judge emails and subject lines by clicks, conversions and revenue instead.
• Build engaged segments on real actions like clicks and purchases, not opens.