Open rates are the wrong metric. Here’s what to watch instead

The number everybody asks about first

Almost every conversation about email marketing starts with open rate. It’s the first number in every platform’s dashboard, it’s the one people compare against benchmarks, and it’s the one that gets quoted in meetings.

It is also, for a business like yours, close to useless. Two separate reasons: it stopped being accurate, and even when it was accurate it wasn’t measuring anything you get paid for.

How an open is actually counted

There is no such thing as an email that tells the sender it was read. The way open tracking has always worked is that the message contains a tiny invisible image hosted on the sender’s server. When a mail app loads that image, the server records a hit and calls it an open.

That mechanism was always rough. Somebody who reads your whole email with images off counts as not opening. Somebody whose preview pane loads the image while they delete it counts as opening. It was a proxy, and everybody agreed to treat the proxy as the thing.

Apple broke the proxy in 2021

In September 2021, with iOS 15, Apple introduced Mail Privacy Protection. When it’s turned on, Apple Mail pre-loads the images in incoming messages through its own proxy, whether or not the person ever looks at the email.

The effect on tracking is direct: every message delivered to a protected Apple Mail user registers as opened. Not some. Effectively all of them. Apple documents the behavior itself, and the prompt asking users to enable it appears the first time they open Mail after updating.

Apple Mail is one of the most common ways people read email, so a meaningful slice of any list is now reporting opens that did not happen. Your open rate went up in late 2021 and the increase was noise.

Mail Privacy Protection stops senders from using invisible pixels to collect information about the user.

Apple, Mail Privacy Protection support documentation

And even a true open rate wouldn’t tell you much

Suppose the tracking still worked perfectly. An open would mean somebody’s eyes were briefly on a subject line. It would not tell you whether they read the piece, remembered your name, thought better of you, or called you six months later.

For a business that gets hired occasionally rather than constantly, the entire value of a newsletter arrives on a delay. It shows up as being the name somebody thinks of when they finally need you. No metric collected at the moment of sending can see that.

What to watch instead

The good news is that the metrics that matter are ones you already have, and none of them require a dashboard.

• Repeat work. How many clients came back this year who had gone quiet? Count them against last year. This is the number the newsletter is most directly responsible for.
• Referrals, and where they came from. Ask every new inquiry how they heard about you and write the answer down. "So-and-so passed on your newsletter" is the highest-value sentence in this business.
• Replies. Actual human responses to the issue. A newsletter that gets five real replies is outperforming one that gets a thousand phantom opens. Make sure replies go somewhere a person reads.
• Mentions in the room. Somebody bringing it up at a meeting, at a closing, in the chair. Unmeasurable and extremely informative.
• Unsubscribes and returned mail. Small numbers, but a rising trend is a genuine signal that something is off with the content or the list.

The digital metrics that survived

Not everything in the dashboard is broken. A few numbers still mean what they say, because they require a deliberate human action.

• Clicks. Somebody chose to click. Bots and scanners create some noise, but a click is still evidence of intent in a way an open is not.
• Read depth on the flipbook. The digital edition reports how far through the issue people actually got. Page three versus page one is a real distinction and it tells you which content earned attention.
• Deliverability. Whether the thing arrived at all. Boring, and far more consequential than open rate.

The one question worth asking every new inquiry

"How did you hear about us?" is the most valuable piece of marketing measurement available to a small business, and most people either don’t ask it or don’t write the answer down anywhere durable.

Ask it on every intake call and every form, and record it somewhere you can count later — a CRM field is ideal, a spreadsheet is fine, a note in somebody’s head is not. The answers are messy and often vague, and that’s all right. Over a year the pattern is legible even when any single answer isn’t.

One caution: attribution from this question undercounts newsletters badly. People say "a friend recommended you" without knowing why the friend thought of you that week. When you hear referral, it is worth one more question about who, because that is where the newsletter usually shows up.

It also stopped being comparable

Because adoption differs by audience, the distortion isn’t even. A list of iPhone-heavy consumers is inflated far more than a list of corporate Outlook users, so comparing your open rate to an industry benchmark now compares two numbers that were measured differently. Your own history is broken across 2021 as well — this year’s open rate and 2019’s are two different definitions of the word.

Somebody will show you an industry average and ask why you are below it. Before answering, check three things: when the figure was published, whether it predates 2021, and how the list was defined.

Benchmarks aggregate wildly different senders. A retailer emailing recent buyers weekly and a law firm emailing past clients monthly appear in the same average and have nothing in common. Comparing across that gap produces a number that is precise and meaningless.

The comparison that is worth making is against yourself. Your list, your content, last quarter against this one. That comparison is at least measuring the same thing twice.

A simple scorecard you can keep on one page

Once a quarter, fill in five lines. It takes ten minutes and it will tell you more than any email platform will.

1. New inquiries this quarter, and how each one said they found you.
2. Clients who returned after a gap of a year or more.
3. Referrals received, and from whom.
4. Replies and direct comments about the newsletter.
5. Unsubscribes and returned mail.

Deliverability is the number that deserves the attention

If you are going to obsess over one figure in the email dashboard, make it the one that says whether your messages arrived. An email that lands in spam has an open rate of zero and a value of zero, and unlike open rate, this is a problem you can actually fix.

Three things do most of the work here. Authenticate your sending domain, which is a one-time technical setup and the single biggest lever. Send from a real domain you own rather than a free mailbox address. And keep the list clean — repeatedly mailing dead addresses is one of the clearest signals a mail provider uses to decide you are not worth delivering.

It is unglamorous work and nobody puts it in a report, which is roughly why it goes undone. It also determines whether every other number on the page means anything.

What we report, and what we don’t

We’ll show you the flipbook read-depth and click data, because those reflect real behavior. We’ll show open rate if you want it, with the caveat above attached, because it’s not worthless — the trend still has some signal even if the level doesn’t.

What we won’t do is build a case for the work out of a number we know is inflated. If a newsletter isn’t producing repeat work and referrals after a fair run, a good open rate is not a defense of it, and we’d rather talk about what to change.