Reading your email metrics

Every email platform gives you a report after each campaign. It's easy to fixate on the first number you see, usually the open rate, and miss the ones that tell you whether email is actually helping your business.

This guide explains each common metric, how much to trust it, and what to do when a number looks off.

Why open rates are unreliable now

An "open" is recorded when an email loads a tiny invisible image, often called a tracking pixel. When the image loads, the platform counts an open.

In 2021 Apple introduced Mail Privacy Protection in its Mail app on iPhone, iPad and Mac. When a user turns it on, Apple loads remote content, including tracking pixels, privately in the background, whether or not the person reads the email. Apple says this hides the reader's IP address and stops senders from learning whether they opened the email.

The result: for subscribers using Apple Mail with this setting on, your platform records an open even if they never looked. Open rates go up, and they no longer tell you much about real interest.

Other things also distort opens:

Security scanners at some companies open emails and click links to check them.
Image blocking in some email apps hides real opens.

Most platforms now label or filter "machine opens" where they can. That helps, but treat open rate as a rough signal at best.

What this breaks

Subject line tests that judge by opens. Judge them by clicks instead.
"Inactive" segments based on opens. Some people who never open look active, and your cleanup misses them. Use clicks or purchases over a longer period.
Send-time optimization that relies on open times.

The metrics that matter

Metric	What it means	How much to trust it
Delivered	The receiving server accepted the email	Reliable, but accepted doesn't mean inbox
Click rate	The share of delivered emails where someone clicked a link	Fairly reliable; security scanners can inflate it a little
Click-to-open rate	Clicks divided by opens	Weak now, because opens are inflated
Conversions	Sales, bookings or form fills that followed a click	The most important; depends on tracking being set up
Unsubscribe rate	People who opted out from this email	Reliable; watch for spikes
Spam complaints	People who clicked "report spam"	Reliable and critical
Bounces	Emails that couldn't be delivered	Reliable; hard bounces mean bad addresses
Replies	People who wrote back	Not always shown, but a strong sign of engagement

Spam complaints: the number that can hurt you

Spam complaints matter more than any other metric because mailbox providers use them to decide where your future email goes. Google's sender guidelines ask senders to keep their spam rate, as reported in Google's Postmaster Tools, below 0.3%, and recommend staying under 0.1%. Yahoo gives a similar 0.3% threshold.

Your email platform shows its own complaint count, but not every mailbox provider reports complaints back, so the platform figure can be lower than reality. If you send a lot to Gmail addresses, set up Google Postmaster Tools for your sending domain to see Gmail's view. See staying out of spam for what to do if the rate climbs.

Track what happens after the click

Your email platform knows someone clicked. Your website analytics knows what they did next. Connect the two:

Add UTM parameters to your email links (many platforms do this automatically if you switch it on), so GA4 shows email as a source. See UTM links for campaign tracking.
Set up key events in GA4 for the actions that matter, such as purchases, form submissions and bookings. See conversions explained.
For stores, connect your email platform to your store. Klaviyo, Mailchimp and others can attribute orders to emails directly.

Be aware that platforms use their own attribution rules. An email platform may count a sale if someone clicked within a few days, even if they also clicked an ad. GA4 may credit a different channel. Neither is wrong; they're answering different questions. Pick one source for decisions and stick to it.

Compare against yourself

Industry benchmark reports exist, but they mix very different businesses and lists, and they're affected by the open-rate problem too. A more useful habit:

Keep a simple log of each campaign: date, subject, list size, clicks, conversions, unsubscribes, complaints.

Look at trends over three to six months, not single sends.

Change one thing at a time (subject style, send day, length, offer) so you know what made the difference.

Watch for sudden changes. A drop in clicks across all emails, or a jump in bounces, often points to a deliverability problem rather than bad content.

What to do when a number looks wrong

Clicks dropped suddenly across every email. Check whether you've started landing in spam. Send a test to a Gmail and an Outlook address and see where it lands.
Unsubscribes spiked after one email. Look at what was different: frequency, topic, a hard sell. One spike isn't a disaster. A pattern is.
High clicks but no sales. The email is working; the landing page may not be. See sending ads to the right page, which applies equally to email.
Hard bounces are high. Your list has bad addresses. Check where they came from, and consider double opt-in.

Frequently asked questions

Should I stop looking at open rates entirely?

You don't need to ignore them, but don't make decisions on them. A large sudden fall in opens can still hint at a deliverability problem.

Does Mail Privacy Protection affect clicks?

No. It affects the tracking pixel that records opens. Clicks are recorded when a person (or a security scanner) follows a link.

What's a good click rate?

It varies so much by business and list that a single figure isn't useful. Track your own and aim to improve it.