Email Marketing Metrics That Matter (and the Ones That Don't)

Email marketing metrics are the numbers that describe how a list responds to a send, and the ones that matter are the ones that would change your next decision. That is the whole test. A metric earns a place in your report only if a different value would make you do something different — change the subject line, fix the landing page, clean the list, or send again. Everything else is decoration. This guide separates email marketing metrics into three layers (delivery, engagement, outcome), shows how to calculate each, explains why opens became unreliable after Apple Mail Privacy Protection, and gives you a reporting routine you can run monthly.

What email marketing metrics actually measure

At the top level, email marketing metrics measure three different things: whether the message arrived, whether anyone acted on it, and whether it produced value. Those are not variations of the same idea — they are separate questions, and mixing them is the most common reason a report looks fine while the program is quietly failing.

A campaign can show a healthy open rate and still be a disaster if complaints are climbing. A campaign can show a low click rate and still be worth sending if it drove a handful of high-value purchases. The number alone tells you nothing until you know which layer it belongs to and what decision it informs.

For a wider view of the market, our ranking of the Best Email Marketing Platforms shows what each service really costs for your number of contacts.

The three layers of email KPIs

Delivery metrics tell you whether the message had a chance. If it never arrived, nothing downstream is meaningful.

Engagement metrics tell you whether it landed. Clicks, replies, and (approximately) opens.

Outcome metrics tell you whether it was worth sending. Conversions, revenue, list growth.

Most reporting problems come from mixing layers. A high open rate on a badly delivered campaign is noise. A low click rate on a campaign that still drove revenue is not a failure — it may be exactly what a small, high-intent segment looks like. Report each layer separately, and read them in order: delivery first, then engagement, then outcome.

Delivery metrics: the floor, not the goal

These are hygiene metrics. They rarely impress anyone, but a rising bounce or complaint rate is an early warning about list quality or sending reputation.

• Delivery rate = delivered ÷ sent. The share of your send that reached an inbox.
• Bounce rate = bounced ÷ sent. Split hard bounces (permanent — the address does not exist) from soft bounces (temporary — a full mailbox, a greylisted server). Hard bounces should never be retried.
• Complaint rate = complaints ÷ delivered. The share of recipients who marked you as spam. This is the metric that gets domains blocked, so watch it closely.
• Unsubscribe rate = unsubscribes ÷ delivered. A steady, low unsubscribe rate is generally not a concern. A spike usually means a mismatch between what people signed up for and what you sent.

Mailerflow's analytics cover delivery, bounces and complaints per campaign and overall, and bounces and complaints are suppressed automatically — which keeps the numbers clean and stops you from mailing addresses that will only hurt your reputation.

Engagement metrics: clicks, replies, and the open-rate problem

• Click-through rate (CTR) = unique clicks ÷ delivered. How many recipients clicked at least one link.
• Click-to-open rate (CTOR) = unique clicks ÷ unique opens. How many of the people who opened also clicked.
• Reply rate = replies ÷ delivered. Relevant mainly for conversational or one-to-one style sends.

Then there is opens. Apple Mail Privacy Protection, introduced in 2021, pre-fetches remote images through Apple's proxy. That means the tracking pixel fires whether or not the recipient ever read the message — an open can be registered the moment the email lands, or never, depending on the recipient's settings. Open rate is now an approximate signal: useful for spotting trends over time, unreliable for absolute judgments or for picking A/B test winners. Mailerflow reports approximate opens, which is the honest framing, and opens are approximate on any platform, so treat exact open figures with caution.

Outcome metrics: what the business actually buys

These are the email KPIs worth reporting upward. They require linking email to a downstream event — a purchase, a signup, an activation.

• Conversion rate = conversions ÷ delivered. Define "conversion" per campaign type before you measure it.
• Revenue per email sent = revenue ÷ emails delivered. The single most useful number for promotional sends.
• Revenue per subscriber = revenue ÷ list size. Good for comparing list segments or acquisition sources.
• List growth rate = (new subscribers − unsubscribes − bounces) ÷ list size. Tells you whether the list is actually growing after churn.

Mailerflow's analytics cover clicks per campaign and overall, and events and webhooks can feed downstream conversion data into your own reporting — so you can join email activity to whatever happens in your store or product.

How to calculate each metric

Here is a compact reference. Every formula uses the same three or four inputs, so once you have sent, delivered, clicks, and conversions, you can compute all of them.

Now a worked example. These numbers are illustrative, not a benchmark — treat them as a way to see how the formulas behave.

Say you send 10,000 emails. 9,800 are delivered, 200 bounce. 2,400 approximate opens register, 240 unique clicks, 24 conversions, and $1,200 in revenue.

• Delivery rate = 9,800 ÷ 10,000 = 98%
• Bounce rate = 200 ÷ 10,000 = 2% (high enough to investigate the source)
• CTR = 240 ÷ 9,800 = 2.4%
• CTOR = 240 ÷ 2,400 = 10%
• Conversion rate = 24 ÷ 9,800 = 0.24%
• Revenue per email = $1,200 ÷ 9,800 = $0.12

What each number tells you: the delivery rate is fine, the bounce rate is worth a look, the CTR is unremarkable, the CTOR says the body copy is doing its job once people open, and the revenue per email is the number your finance lead actually cares about. Notice that the open count is only used once, in CTOR — and even there, it is approximate.

Why opens became unreliable

Apple Mail Privacy Protection loads remote images through Apple's proxy servers. Because the tracking pixel is an image, it fires whether or not the recipient read the message. The practical consequences are threefold.

First, open rates are inflated and inconsistent. Some recipients always register an open; some never do. You cannot compare your open rate to a published figure, and you cannot compare two periods if the mix of Apple Mail users changed.

Second, open-based segments stop meaning "interested." A segment of "people who opened the last three emails" now includes people who never looked at any of them. If you use that segment for re-engagement or suppression, you are acting on bad data.

Third, subject-line A/B tests can produce false winners. If opens are noisy, a test that declares a winner on open rate may just be measuring which pixel loaded more often.

Practical replacements: test on clicks or conversions instead of opens. Segment on click and purchase behaviour rather than opens. Treat open rate as a directional trend, not a verdict. Mailerflow labels opens as approximate rather than presenting them as exact, which is the right way to report them.

Choosing the metrics that match your goal

Where Mailerflow helps here: embeddable signup forms with double opt-in and consent records make signup conversion measurable rather than guessed, and dynamic segments on engagement, events and website behaviour let you build audiences from clicks and actions rather than opens.

A worked example: reading one campaign's numbers

Take the illustrative campaign from earlier and walk it through all three layers.

Delivery layer: 98% delivered, 2% bounced. The bounce rate is the first thing to act on. A bounce rate this high can point to a specific acquisition source — a purchased list, a co-registration partner, or a form that is not validating addresses. Check where those 200 addresses came from before you send again.

Engagement layer: 2.4% CTR, 10% CTOR. The CTOR is healthy, which tells you the offer and body copy are working for the people who see the email. The CTR is modest, which suggests the audience is small or the subject line is not earning the open. Because opens are approximate, do not A/B test the subject line on open rate — test it on clicks.

Outcome layer: 0.24% conversion, $0.12 revenue per email. If your average order value is high, this may be a perfectly good send. If it is low, the conversion rate is the problem, and the fix is almost certainly the landing page, not the email. A healthy CTR with weak conversion is a landing-page signal, not an email signal.

Common mistakes when reporting email performance

1. Reporting open rate as a headline number. After MPP, it is approximate. Lead with clicks, conversions, or revenue instead, and mention opens only as a trend.
2. Comparing CTR across lists of different sizes without normalising. For example, a 5% CTR on a small list and a 5% CTR on a much larger list are not the same achievement. Report rates, not raw counts.
3. Using CTOR when opens are unreliable. CTOR depends on the open count, so it inherits the same noise. Use it as a secondary diagnostic, not a primary metric.
4. Ignoring unsubscribes and complaints because they are small. If a complaint rate doubles from, say, 0.02% to 0.04%, it is still small, but it is a doubling. Watch the trend, not the absolute value.
5. Measuring a campaign in isolation. Compare a promotional send to your other promotional sends, not to your onboarding sequence. Build a baseline per campaign type.
6. Treating a single send as a verdict. One campaign is a data point. Three campaigns with the same pattern is a signal. Wait for the pattern.

Checklist: a reporting routine you can run monthly

• Verify delivery and complaint rates first. If these are off, nothing else matters.
• Track click and conversion trends per campaign type, not per individual send.
• Report revenue per email or per subscriber where you can attribute it.
• Review list growth and unsubscribe rate together — growth without retention is churn in disguise.
• Re-check any open-based segment or A/B test decision, since opens are approximate.
• Keep a short note of what you changed and what happened, so next month's report has context.

Mailerflow's per-campaign and overall analytics cover delivery, bounces, complaints, clicks and approximate opens, which handles the first two steps without extra tooling. If you want to see how that looks on your own list, the features page walks through the analytics and segmentation side, and the pricing page shows what is included on each plan.

FAQ

What is a good open rate now?

There is no reliable universal figure after Apple Mail Privacy Protection, because opens are inflated inconsistently depending on how many of your recipients use Apple Mail. Compare against your own history instead: is the trend up or down over the last six months? That is the only comparison that means anything.

What is the difference between CTR and CTOR?

CTR is unique clicks divided by delivered — it measures how many recipients acted. CTOR is unique clicks divided by unique opens — it measures how many of the people who opened also clicked. CTOR is useful for judging body copy, but because it depends on the open count, treat it as a secondary diagnostic.

Which email metrics should I report to leadership?

Lead with outcome metrics: revenue per email, conversion rate, and list growth rate. Add delivery health (bounce and complaint rates) as a one-line risk indicator. Keep engagement metrics (CTR, approximate opens) as supporting detail rather than the headline.

How do I measure email revenue if my store is separate?

You need to link email activity to a downstream event. The usual approach is UTM parameters on your links plus a conversion event fired on your store, or a webhook that pushes order data back into your email platform. Mailerflow supports signed webhooks and events, so you can join email clicks to purchases in your own reporting.

What is a healthy unsubscribe rate?

There is no universal number, but the trend matters more than the absolute value. A steady, low unsubscribe rate is generally not a concern — people's interests change. A sudden spike usually means the content did not match what people signed up for, so check the acquisition source for that send.

Put this into practice in Mailerflow: create your first campaign or automation by telling the AI what you want. Start free.