Email Marketing Metrics Explained: Formulas, Limits and What to Trust

A campaign report is a handful of counts and the rates between them. To read one well, you need to know two things about every rate: what it divides by, and what can make the count on top wrong. This guide goes through both, metric by metric.

This guide covers:

1. The five counts behind every report
2. Delivery: delivered and bounced
3. Engagement: opens, clicks and unsubscribes
4. Why opens and clicks can mislead
5. Complaints: the number providers watch
6. Outcomes: conversions, revenue and ROI
7. Reading a campaign report in AcelleMail
8. Which numbers to judge a campaign by

It ends with the most common mistakes and a one-screen summary.

1. The five counts behind every report#

Every rate in a campaign report is one of five counts divided by another:

Two conventions decide whether two numbers can be compared:

• Unique or total. A unique count counts each person once, however many times they opened or clicked. A total count includes every repeat. Rates use unique counts; a total is useful only to see how often people came back.
• What the rate divides by. Most tools divide opens and clicks by delivered, some by sent. The same campaign can show two different open rates in two tools. Check before you compare.

2. Delivery: delivered and bounced#

Metric	Formula	What it tells you
Delivered	sent − bounced	How many messages the receiving servers accepted. It says nothing about inbox or spam folder.
Delivery rate	delivered ÷ sent	The share that was accepted.
Bounce rate	bounced ÷ sent	The share that was refused. Split it by type before you act on it.
Hard bounce rate	hard bounces ÷ sent	The quality of your list. On a list collected with consent and cleaned regularly, it stays close to zero.

A hard bounce is permanent (the address does not exist); a soft bounce is temporary (a full mailbox, a busy server). Sending services judge you on hard bounces. Amazon SES, for example, counts "only hard bounces", asks senders to stay below 2%, puts an account under review at 5% and may pause sending at 10% (Amazon SES sending review FAQ). Other services publish their own limits.

A high delivery rate does not mean your mail reached the inbox. That second decision is covered in Email Deliverability Explained; what to do about hard bounces is in Email List Quality Explained.

3. Engagement: opens, clicks and unsubscribes#

Metric	Formula	What it tells you
Open rate	unique opens ÷ delivered	Roughly how many people saw the email. Useful for trends in your own sends, weak as an absolute (section 4).
Click rate	unique clicks ÷ delivered	How many people acted on the email. The most dependable engagement signal you have.
Click-to-open rate (CTOR)	unique clicks ÷ unique opens	How well the content persuaded the people who opened. Automatic opens push it down.
Unsubscribe rate	unsubscribes ÷ delivered	How many people decided they no longer want your mail. A jump after one campaign points at its content or its timing.

How the counts are collected matters for what follows. An open is recorded when a tiny image in the email, a tracking pixel, loads from your server. A click is recorded when a link first passes through a tracking address that notes the click and then forwards the reader. In AcelleMail both are switched on per campaign, with Track opens and Track clicks under Advanced Settings in the campaign's Setup step. With either switched off, its rate stays at zero.

4. Why opens and clicks can mislead#

Both signals depend on software behaving like a reader, and three things break that link:

Apple Mail Privacy Protection. With it on, Apple Mail loads the images in an email when the message arrives: remote content "is privately downloaded in the background when you receive a message (instead of when you view it)" (Apple, Mail on Mac). In Apple's words the feature "prevents senders from seeing if you've opened the email message they sent you" (Apple, Mail on iPhone). Every Apple Mail reader who turned it on counts as an open, read or not. The share varies with your audience, so the same open rate can mean different things on two lists.

Images turned off. Some mail apps, and some readers, load images only when asked. Those people can read the whole email and never count as an open. A click still counts, so a reader who clicked has read the email whether or not an open was recorded.

Link scanners. Security filters in many company mail systems follow the links in incoming mail to check them. Microsoft Defender for Office 365, for one, says "URLs are scanned prior to message delivery" (Microsoft). A scan can register as a click. The pattern gives it away: every link in the email clicked within seconds of delivery, often from the same company domain.

What this means in practice:

• Judge a campaign's success by clicks and what came after them, not by opens.
• Use the open rate to compare your own sends with each other: the same list, a similar time, two subject lines in an A/B test.
• Look past opens when you decide who has gone quiet; clicks are the better sign of a reader who left. See Email List Quality Explained, section 6.

5. Complaints: the number providers watch#

The complaint rate (also called spam rate) is complaints ÷ delivered: how many people pressed "Report spam". It is the one engagement number mailbox providers publish limits for:

• Gmail asks senders to "keep spam rates reported in Postmaster Tools below 0.10% and avoid ever reaching a spam rate of 0.30% or higher" (Email sender guidelines). It counts only mail that reached the inbox: the spam rate "is the percent of your messages that are delivered to engaged recipient's Inbox and then marked as spam by the recipient".
• Yahoo asks bulk senders to stay below 0.3% (Sender best practices).
• Amazon SES puts an account under review at 0.1% and may pause sending at 0.5%, counting only "mail sent to domains that send complaint feedback to SES".

Why your report shows fewer complaints than there are. A sending platform hears about a complaint only when the mailbox provider reports it back through a feedback loop. Gmail does not report individual complaints at all; its spam rate is visible only in Gmail Postmaster Tools. A campaign report that shows zero complaints means none came back, not that nobody complained.

6. Outcomes: conversions, revenue and ROI#

Metric	Formula	What it tells you
Conversion rate	conversions ÷ delivered, or ÷ unique clicks	How many recipients did what the email asked: bought, booked, signed up. Always say which denominator you used.
Revenue per email	revenue from the campaign ÷ delivered	What each delivered email earned. Lets you compare campaigns of different sizes.
Return on investment (ROI)	(revenue − cost) ÷ cost	Whether the campaign paid for itself.
List growth rate	(new contacts − contacts lost) ÷ contacts at the start of the period	Whether the list grows faster than it loses unsubscribes, bounces and complaints.

None of these happen in the email, so your email platform cannot see them on its own. Tag your links so your analytics tool or shop can tell which campaign sent the visitor. In AcelleMail, switch on Enable UTM tracking under UTM tracking in the campaign's Setup step; every link then carries the campaign's UTM parameters. Details: UTM Parameters in AcelleMail and Email Campaign ROI.

A worked example#

The counts from section 1, with the bounces and complaints added. Illustrative numbers, not a benchmark:

Result	Count
Sent	10,000
Bounced	200 (150 hard, 50 soft)
Delivered	9,800
Unique opens	3,920
Unique clicks	490
Unsubscribes	29
Complaints	5
Orders	49

Metric	Calculation	Result
Delivery rate	9,800 ÷ 10,000	98%
Bounce rate	200 ÷ 10,000	2% (hard: 1.5%)
Open rate	3,920 ÷ 9,800	40%
Click rate	490 ÷ 9,800	5%
Click-to-open rate	490 ÷ 3,920	12.5%
Unsubscribe rate	29 ÷ 9,800	0.30%
Complaint rate	5 ÷ 9,800	0.05%
Conversion rate (of delivered)	49 ÷ 9,800	0.5%
Conversion rate (of clicks)	49 ÷ 490	10%

The complaint rate looks safe at 0.05%. Remember that it counts only the complaints that came back to you.

7. Reading a campaign report in AcelleMail#

Open Campaigns → Email and click a sent campaign. Its report starts with four cards:

1. Recipients: every contact on the lists or segments the campaign was sent to. That includes the contacts it skipped because they were not Subscribed, or because email verification found them Undeliverable, Risky or Unknown. It is not the number sent. In this example, 1,938 contacts were on the list and 1,207 of them were Subscribed.
2. Delivered: the messages your sending server accepted, 1,187 here, including any that bounce later. Its percentage is of Recipients, so it is not a delivery rate. Messages that could not be sent at all show as Failed under Statistics breakdown, 20 in this case. The delivery rate of what was actually sent is 1,187 ÷ (1,187 + 20) = 98.3%.
3. Opened: unique opens, as a percentage of Delivered.
4. Clicked: unique clicks, as a percentage of Delivered. The click-to-open rate is not shown as a figure of its own; here it is 47 ÷ 469 = 10%.

Further down, the Click rate card lists Total clicks next to Unique clicks. The rate uses the unique count.

The figures are saved, not live. At the top of the report, next to "Last updated", the ↻ button (Refresh campaign data) recalculates them in the background; come back a few minutes later. In the campaign list, a campaign whose figures need updating says "No statistics or out-dated, click to update".

Who did what. The Sending logs menu on the report (Logs on an A/B test) lists each contact behind the numbers: the Delivery log, Bounce log, Feedback log, Open log, Click log and Unsubscribe log.

Comparing campaigns. The campaign list shows Delivered, Open rate and Click rate for every campaign. Type part of a name into the search box to line up similar sends:

The number after the slash in the Delivered column is the same Recipients count as on the report: every contact on the list.

8. Which numbers to judge a campaign by#

• Success: clicks and conversions. What readers did.
• Health: complaints, hard bounces and unsubscribes. Whether your list and your reputation are in good shape. These are the numbers with limits attached (section 5).
• Opens: trends only. Compare within your own sends, never as an absolute.

This guide gives no "industry average" on purpose. Published averages mix audiences, countries and counting methods, and many cannot say how much of their open rate is automatic. Your own history is the benchmark that matters: the same kind of campaign, to the same list, over time. When you want to know whether a change helped, change one thing at a time; see A/B Testing Email Subject Lines.

9. Common mistakes#

What you see	Why it is a problem	What to do
Your open rate compared with an "industry average" or another tool's	Counting methods and denominators differ	Compare with your own earlier campaigns
Recipients read as "how many we sent"	It includes every contact the campaign skipped	Use Delivered plus Failed for what was sent
A rising open rate celebrated	Automatic opens can rise on their own, with no new readers	Check that clicks rose too
Total clicks used as the click count	Repeat clicks and link scanners inflate it	Use unique clicks
"Zero complaints" taken at face value	Gmail does not report complaints back	Check Gmail Postmaster Tools
A report full of zeros right after sending	The figures are saved and not yet updated	Click ↻ next to "Last updated"
A conversion rate with no denominator	Of delivered and of clicks differ tenfold	Say which one you used

Summary#

1. Every rate is one count divided by another. Know which, and use unique counts.
2. Delivered means accepted by the receiving server, not placed in the inbox.
3. Opens are skewed both ways: Apple Mail Privacy Protection adds opens, blocked images hide them. Use them for trends only.
4. Judge success by clicks and conversions, and watch for link scanners.
5. Keep complaints below 0.1% and hard bounces near zero. Your report sees only the complaints that come back to you.
6. Tag links with UTM parameters to measure what happens after the click.
7. In AcelleMail, Recipients counts every contact on the list; read rates against Delivered, and refresh saved figures with ↻.
8. Benchmark against your own campaigns, not against averages.

Terms used here are defined in the Email Marketing Glossary.

♡ Sign in to like 0 💬 0 comments