What is a good email open rate for ecommerce?

An open is recorded when a tiny image in the email is loaded. That was always approximate, because plain text views and blocked images went uncounted. It became far less reliable once mail applications began pre-loading images on behalf of the recipient, which registers an open whether or not a person ever looked at the message.

The result is a number inflated by an unknown amount that differs by audience. A list weighted towards one mail platform will show a higher rate than an identical list weighted towards another, with no difference in behaviour.

So comparing your rate to a published figure is worse than useless, because the published figure was collected from a different mix of mail clients.

What the number can still do is compare your own sends with each other, sent to similar segments in a similar period. If one subject line produces a noticeably different rate than another across a large enough audience, that difference still carries some information, since the inflation applies to both.

And a sudden change in the rate is worth investigating, not because the level matters but because a sharp fall often means messages are being filtered rather than ignored. That is a deliverability signal, and it is the most useful thing the metric still provides.

Four measures survive the change and connect to money.

Click rate, measured against delivered messages. A click is a person doing something, which is harder to fake and harder to inflate. Watch it per send and per flow.

Revenue per recipient. Take the revenue attributed to a send and divide by the number of people it reached. This single figure makes sends comparable regardless of list size, and it stops the habit of judging email by total revenue, which simply rewards sending more often to more people.

Unsubscribe rate and spam complaint rate. These are the cost side of every send. A campaign that made money this week while burning through the list has borrowed from next quarter, and complaints in particular damage deliverability for every future message.

List growth against list decay. How many new subscribers arrive, how many leave, and what share of the list has engaged in the last few months. A list that grows while engagement falls is heading for filtering problems.

Together these tell you whether the programme is healthy. Opens tell you almost nothing on their own, and the stores that improved most in the last few years are the ones that stopped optimising for them.

Reaching the inbox is a bigger factor than the subject line, and it is decided before anyone reads a word.

Authentication comes first. The technical records that prove you are allowed to send from your domain need to be set up correctly, and they need to stay correct when you change tools or domains.

Sending reputation comes next, and it is earned by sending to people who want the messages. Sending to everyone regardless of engagement is the reliable way to lose the inbox for the people who do care.

So segment by engagement. Send the regular schedule to the people who have opened or clicked recently, and put the long silent group on a much lighter schedule or a sunset sequence. Removing dead addresses feels like shrinking the asset. It is maintaining it.

List collection matters too. Addresses gathered from a competition, a bought list or an aggressive popup produce complaints, and complaints follow you.

Then the subject line does its job: telling a person who already wants to hear from you what this message contains. Clarity beats cleverness, because a subject that sets an accurate expectation earns clicks and the next open too.

And timing follows the product, not a chart of best send times. Reminders belong when the product runs out, not on a Tuesday morning.

Three habits keep email reporting honest.

Compare flows and campaigns separately. Automated flows reach people at a moment of intent, so their rates are much higher, and blending them with a newsletter produces an average that describes nothing. Report them apart, always.

Segment before you conclude. A send to your most engaged customers and a send to the whole list are different activities. If the split changes month to month, every headline rate changes with it.

Judge over a period, not per send. Email revenue arrives unevenly, and one strong campaign can hide two weak months. A rolling view of revenue per recipient across all sends is the number that tells you whether the programme is improving.

One more thing worth checking. Attributed revenue in an email tool and revenue in the store rarely match, because the tool credits orders within a window after a click or an open. Understand the window your tool uses before quoting the figure to anyone, and sanity check email's contribution against total store revenue over the same period.

Set a small set of numbers, look at them monthly, and resist adding more. Email reporting fails more often from too many metrics than from too few.