# Email Open and Click Rates

> What open, click and unsubscribe rates tell you. Why open rates have looked too high since a privacy change in 2021.

**Course:** [Email Marketing](https://learn.benibix.com/learn/marketing-email)  
**Time:** 11 min

**After this lesson you can:**

- Work out open and click rates
- Explain why open rates look too high
- Judge an email test by clicks or sales

## Which email numbers to trust

Three numbers describe how an email did. Open rate is the share of delivered emails that were opened. Click-through rate is the share that got a click. Unsubscribe rate is the share of people who left. Since September 2021, open rates have looked too high, because Apple Mail Privacy Protection opens emails by itself. Clicks are the number to trust.

Opens are counted by a tiny hidden picture in the email. When the picture loads, the email counts as opened. Apple Mail Privacy Protection loads each picture by itself, before the person reads a word. So each email to a reader with it turned on can look opened.

That makes opens a rough guide at best. A jump in open rate may only mean more of your readers have the setting on. Tests that pick a winner by opens, and resends to people who did not open, now work on bad data.

Clicks are harder to fake. A click means a real person wanted to see more. Divide clicks by delivered emails for the click-through rate. Some tools also show clicks divided by opens, and that has the same problem.

Some people leave after every send. That is normal, and it keeps the list healthy. A sudden rise after one email tells you something in it missed. It might be the subject, the timing or the offer.

An A/B test sends two versions of an email to two small groups and compares them. Change one thing at a time, such as the subject line or the button. Pick the winner by clicks or sales, then send it to the rest of the list.

The common mistake is comparing your open rate with one from before 2021, or with another brand. Lists differ in how many readers have the setting on. Compare an email with your own past emails, and use clicks.

## The three email numbers

| Number | How it is worked out | How far to trust it |
| --- | --- | --- |
| Open rate | Opens divided by delivered emails | A rough guide, pushed up by privacy settings |
| Click-through rate | Clicks divided by delivered emails | The most reliable of the three |
| Unsubscribe rate | People who left divided by delivered emails | Watch for sudden jumps |

![Four bars for one email sent to 10,000 people. Delivered: 9,800. Opened: 4,100, with the top half shaded to show opens counted by Apple Mail Privacy Protection that may not be real reads. Clicked: 290. Unsubscribed: 20. A note says judge by clicks, not opens.](https://learn.benibix.com/diagrams/mkt-email-rates.svg)

### Example: A shop picks the wrong winner

A shop tests two subject lines. Version A gets 44 in 100 opens and 2 in 100 clicks. Version B gets 39 in 100 opens and 4 in 100 clicks.

The tool picks A by opens and sends it to 40,000 people. Judged by clicks, B would have brought about 800 more visits to the shop.

**Words in this lesson:** [Email Open Rate](https://learn.benibix.com/term/marketing/open-rate), [Apple Mail Privacy Protection](https://learn.benibix.com/term/marketing/apple-mail-privacy-protection), [Click-Through Rate (CTR)](https://learn.benibix.com/term/marketing/click-through-rate), [Unsubscribe Rate](https://learn.benibix.com/term/marketing/unsubscribe-rate), [A/B Test](https://learn.benibix.com/term/marketing/ab-test)

**Question:** Why are email open rates less reliable than they used to be?

**Answer:** A privacy setting loads emails by itself, so they look opened even if unread. Since 2021, one setting loads the hidden picture in each email by itself. So the email counts as opened, even if nobody read it.

---

Source: https://learn.benibix.com/learn/marketing-email/email-open-and-click-rates
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-20
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
