Measuring email: which numbers actually matter

Email tools produce a lot of numbers: open rates, click rates, bounce rates, unsubscribe rates, maps of where people click. It is easy to spend time watching them move up and down without learning anything useful. For a membership site owner, the real question is simpler: are your emails helping people join, stay and get value from the membership?

Some email numbers help answer that. Others are less reliable than they look. Here is what each one can and cannot tell you.

Open rates: a rough signal, not a score

The open rate is the share of delivered emails that were opened. It is usually measured with a tiny invisible image in the email; when the image loads, the tool records an open. That method has real limits. Some email apps load images automatically, recording opens from people who never looked at the message, while others block images and hide genuine opens. The result is that open rates can be inflated for some audiences and understated for others.

Opens are still useful for rough comparisons within your own list, such as whether one style of subject line tends to do better than another over several sends. They are not reliable enough to judge individual subscribers, to compare yourself with other businesses or to be a goal on their own. If you do want to improve them, the answer usually lies in writing subject lines people open.

Clicks: evidence of real interest

A click means someone read enough to act. Two measures are worth knowing:

• Click rate: the number of people who clicked, divided by the number of emails delivered.
• Click-to-open rate: people who clicked divided by people who opened. It shows how compelling the email itself was, though it inherits the weaknesses of open tracking.

More valuable than either is looking at which links were clicked. If the forum highlight in your weekly email consistently gets more clicks than the new lesson, that tells you something about what members want.

Conversions: what you actually care about

A conversion is the action an email exists to produce: joining, upgrading, registering for an event, starting a course, updating a card, renewing. Decide the conversion for each email before you send it, then measure that. A launch email that gets fewer clicks but more sign-ups is the better email.

To see conversions, you need to connect the click in the email to what happens on your site. Campaign tags, small labels added to the end of a link that your analytics can read, show which email brought each visitor. Some membership software can also report which members logged in or joined after a given email.

Health numbers: bounces, complaints and unsubscribes

These tell you whether your list and your sending reputation are in good shape:

• Bounces are addresses that could not receive the email. A sudden rise usually points to a bad list import or a problem with your sending setup.
• Spam complaints are the most serious. Even a small number can hurt your ability to reach inboxes, so any jump deserves investigation.
• Unsubscribes are normal, and a steady trickle is healthy. Watch for spikes after particular emails, which suggest a mismatch between what you sent and what people expected.

Worked example: comparing two emails

Imagine the online membership of a yoga studio sends two emails to 2,000 members in the same month. The numbers are purely illustrative:

• Email A, a new class announcement: 900 recorded opens, 60 clicks and 45 members who started the class.
• Email B, a personal note about building a home practice: 700 recorded opens, 140 clicks, 110 members who started the recommended class and 25 replies.

By opens alone, Email A looks better. By clicks, conversions and replies, Email B is clearly stronger: it got more members practicing, which is what keeps them subscribed. The lesson for this studio is to send more emails in the style of Email B and watch whether the pattern holds.

Traps to avoid

• Reading too much into small numbers. With a few hundred subscribers, a handful of clicks either way can look like a big swing.
• Comparing different audiences. Prospects and members behave differently, so compare like with like.
• Changing many things at once. If you change the subject, layout and send time together, you will not know which one mattered. A/B testing for membership sites explains how to test with modest traffic.
• Chasing benchmarks. Your own trend over time means more than someone else's average.

A simple monthly email review

Once a month, spend twenty minutes on the questions below, and resist the urge to track more. Using analytics without drowning in data makes the case for measuring only what informs a decision.

1. Which email produced the most conversions, and why do you think it worked?
2. Which links got the most clicks across all your emails?
3. Did bounces, complaints or unsubscribes spike after any send?
4. Are clicks per email trending up, down or flat over the last few months?
5. What one change will you try before the next review?