Measuring email marketing performance now open rates lie

Aug 27
3 min read

Updated: 2 days ago

Introduction

For years the standard measure of an email campaign was its open rate. That number has become unreliable, and many small businesses are still steering by it.

Privacy features on the major mail platforms pre-load images automatically, registering opens that never happened, while blocking tracking for other recipients entirely. The figure now contains both phantom opens and invisible real ones.

1. Measuring email marketing performance means moving away from open rate

Understand why the number broke before choosing a replacement.

Opens are detected by a hidden image loading. When a mail provider loads that image on every message regardless of whether anyone read it, the open rate becomes a measure of the provider's behaviour rather than your subscribers'.

2. Use clicks as your engagement measure instead

Clicks require a deliberate human action, which makes them far harder to distort.

Click rate, and specifically clicks as a proportion of delivered mail, is now the most reliable indicator of whether the content interested anybody. It is not perfect, but it reflects a decision rather than an automatic process.

3. Track revenue per recipient as the primary figure

For any commercial email, this is the number that matters.

Total revenue attributable to a send, divided by the number of recipients. It captures the whole chain from subject line to purchase in a single figure and lets you compare campaigns that had quite different structures.

4. Count replies, because they indicate real attention

An underused measure and a genuinely informative one.

Somebody who writes back has read the message and cared enough to respond. For service businesses in particular, replies are a better indicator of a successful email than any percentage, and they frequently lead directly to work.

5. Watch unsubscribes and complaints as your warning lights

These two figures tell you when something is wrong.

A rising unsubscribe rate suggests the content or the frequency has drifted from expectations. Spam complaints are more serious and threaten your deliverability. Both should be reviewed on every send, not quarterly.

6. Look at delivery and bounce rates before concluding anything

A campaign that performed poorly may never have arrived.

Check delivered versus sent, hard and soft bounces, and any provider-specific failures. Diagnosing a content problem when the real issue was authentication or list quality leads to rewriting emails that were perfectly good.

7. Compare like with like

Averages across dissimilar campaigns are meaningless.

A transactional message, a welcome email, a newsletter and a promotion have completely different natural engagement levels. Compare each type against its own history, and segmented sends against other segmented sends.

8. Give a change enough data before judging it

Small lists produce noisy results.

A few hundred recipients cannot reliably distinguish a small improvement from random variation. Either accumulate several sends before concluding, or accept that only large differences are detectable at your list size — and say so rather than acting on noise.

9. Attribute revenue honestly, including the delayed purchases

Email frequently influences a purchase that happens later through another route.

Someone reads your email, thinks about it, and buys three days later by searching for you. Last-click attribution credits that to search. Look at overall revenue in the days after a send as well as direct click-through purchases.

Keep a simple record of each campaign, its audience and its result. Six months of that is worth more than any platform dashboard, because it is the only view that reflects your own list.

Conclusion

Stop steering by open rate, because privacy features have made it a measure of mail providers rather than of subscribers.

Use click rate as your engagement indicator, make revenue per recipient the primary figure, count replies as evidence of real attention, treat unsubscribes and spam complaints as warning lights checked every send, verify delivery before diagnosing content, compare each email type against its own history, accumulate enough sends before drawing conclusions from a small list, and account for the delayed purchases that email influences but does not directly close.