Email marketing metrics: the ones that matter (and benchmarks)
This week’s campaign hit a 48% open rate, the best of the quarter. The e-commerce manager celebrated in the team chat. Three days later, finance showed that revenue attributed to email had fallen over the same period. Nobody lied: both numbers were correct. One of them just no longer measures what it appears to measure.
Ever since Apple started automatically opening emails for Mail app users with privacy protection turned on, open rates have been inflated and lost much of their value as a signal of interest. Yet they still sit at the top of most reports. It’s the classic case of measuring what’s easy instead of what matters.
The email marketing metrics that matter are, in this order: deliverability (bounce rate and spam complaint rate), click rate, click-to-open rate (CTOR), conversion and revenue per email sent, plus the unsubscribe rate as a health check for your list. Opens still work as a signal for subject lines and delivery, not as proof of interest.
Why did the open rate stop being the main metric?
Opens are tracked with an invisible pixel in the email body. When the image loads, the tool counts an open. Apple Mail Privacy Protection preloads those images on Apple’s servers whether or not the person read the message. The result: some of your "opens" are made by a machine.
The MailerLite benchmark report, published in December 2025 with data from 3.6 million campaigns, says so explicitly: real open rates are lower than dashboards show, and click rate is the most reliable engagement indicator because it doesn’t depend on open tracking.
That doesn’t mean throwing opens away. They still help in two situations: comparing subject lines within the same list (Apple’s effect hits both versions equally) and spotting delivery problems, since a sudden fall in opens usually means your emails started landing in spam. The mistake is using opens as a campaign goal or as proof that the content worked.
Email marketing metrics in four layers
The most useful way to organize these metrics is by dependency. If the email isn’t delivered, nobody opens it. If nobody clicks, there’s no sale. Each layer only makes sense when the one before it is healthy, and any diagnosis should start at the top.
1. Delivery: bounce rate and spam complaints
Bounce rate = bounced emails ÷ emails sent × 100. Separate hard bounces (the address doesn’t exist) from soft bounces (full inbox, server down). Hard-bounced addresses should leave your list the same day. Retrying them damages your sender reputation.
Spam complaint rate = spam reports ÷ emails delivered × 100. It’s the most underrated metric on this list. Gmail’s sender guidelines ask you to keep it below 0.10% in Postmaster Tools and never let it reach 0.30%. Since late 2025, Gmail has enforced these rules more strictly and may reject messages from non-compliant senders instead of just routing them to spam.
2. Attention: open rate
Open rate = unique opens ÷ emails delivered × 100. Use delivered, not sent, as the denominator so you don’t mix a delivery problem with a subject line problem. Treat it as a relative signal: compare it with your own history and across subject line variants, never as an absolute target.
3. Action: click rate and CTOR
Click rate (CTR) = unique clicks ÷ emails delivered × 100. It measures how many people on your list took action. It follows the same logic as CTR for ads and search, except the audience here is already yours.
Click-to-open rate (CTOR) = unique clicks ÷ unique opens × 100. It isolates content quality: of the people who opened, how many found a reason to click. Because opens are inflated, true CTOR is higher than your dashboard shows, but comparisons between campaigns still hold.
4. Outcome: conversion and revenue per email
Conversion rate = conversions ÷ unique clicks × 100. It depends as much on the landing page and offer as on the email, so read it alongside your site conversion rate.
Revenue per email sent = attributed revenue ÷ emails sent. This number sums up every layer in a unit finance understands. Many teams report it per thousand sends (multiply by 1,000) to avoid fractions of a cent.
List health: unsubscribe rate
Unsubscribe rate = unsubscribes ÷ emails delivered × 100. An unsubscribe isn’t a disaster. It’s someone leaving before they turn into a spam complaint. The warning sign is the trend: if the rate climbs for three or four sends in a row, the issue is usually frequency or relevance, not the copy of one specific email.
Benchmarks: what is a good open rate and click rate?
A benchmark is a starting point, not a goal. It varies by tool, country, list size and even by how opens are counted. That said, it helps to know which range you’re in. The figures below are medians from MailerLite’s customer base between December 2024 and November 2025.
| Industry | Open rate | Click rate | CTOR |
|---|---|---|---|
| All industries | 43.46% | 2.09% | 6.81% |
| E-commerce | 32.67% | 1.07% | 4.01% |
| Software and web apps | 39.31% | 1.15% | 5.40% |
| Marketing and advertising | 37.23% | 1.30% | 6.09% |
| Agencies | 40.52% | 1.85% | 6.74% |
| Higher education | 43.98% | 2.15% | 9.15% |
The median unsubscribe rate in the same study was 0.22% per send. Notice two things. First, e-commerce has the lowest rates in the table because stores send more often: an online store with a 1.5% click rate is above its industry median, even though it’s below the overall average. Second, this is a global dataset made up mostly of small businesses.
In practice, the most useful benchmark is your own history over the last six months, split by email type. Newsletters, promotions and automated flows perform very differently and shouldn’t be blended into a single average.
Example: the campaign that lost on opens and won on revenue
A fashion retailer sends two campaigns to lists of the same size, 20,000 contacts each. Campaign A uses a curiosity-driven subject line with no mention of the offer. Campaign B states the discount and who it’s for right in the subject line.
| Metric | Campaign A | Campaign B |
|---|---|---|
| Emails delivered | 19,600 | 19,700 |
| Unique opens | 9,016 (46%) | 7,486 (38%) |
| Unique clicks | 294 (1.5%) | 552 (2.8%) |
| CTOR | 3.3% | 7.4% |
| Orders | 9 | 22 |
| Revenue | $2,700 | $6,160 |
| Revenue per thousand sends | $135 | $308 |
If the team only looked at opens, they’d repeat the curiosity subject line next time. Campaign B got fewer opens because its subject line already filtered out people who weren’t interested. Whoever opened, opened to buy: CTOR more than doubled and revenue per thousand sends was 2.3 times higher. This pattern is common. Subject lines that spark curiosity without a clear promise inflate opens and starve clicks.
The numbers are a simulation, but the math applies to any list: when opens and revenue point in opposite directions, trust revenue.
How to diagnose problems by reading metrics together
A single metric rarely explains anything. Decisions come from combinations. These are the patterns that show up most often:
| What you see | Likely cause | What to do |
|---|---|---|
| Opens fell suddenly across all campaigns | Emails landing in spam | Check complaints and domain authentication (SPF, DKIM, DMARC) |
| High opens, low clicks | Subject line promises something the body doesn’t deliver | Align subject and content, with one main CTA |
| Good clicks, few sales | Landing page or offer | Review the page, the price and the path to checkout |
| Unsubscribes rising for weeks | High frequency or stale list | Send less to inactive contacts and segment |
| Spam complaints above 0.10% | Contacts who never asked to hear from you | Clean the list and review where sign-ups come from |
This reading order saves a lot of meetings: before debating campaign copy, confirm the email actually arrived. Plenty of teams spend weeks testing subject lines when the real problem is delivery.
Which email marketing metrics belong in the report?
For leadership, three numbers are enough: revenue per thousand sends, click rate and spam complaint rate. The first shows results, the second shows real engagement and the third shows whether the channel is at risk. Opens and CTOR belong in the team’s operational report, where they’re useful for testing subject lines and content.
One technical caveat: the revenue your email tool attributes to itself tends to be generous. Tag your links with UTM parameters and check revenue in GA4 or your e-commerce platform. The two numbers rarely match, and the gap says a lot about each tool’s attribution window. If email is your main repeat-purchase channel, also track your customer retention rate: a healthy email program shows up there before it shows up in monthly revenue.
Frequently asked questions
What is a good email marketing open rate?
In MailerLite’s data, the median between December 2024 and November 2025 was 43.46% overall and 32.67% for e-commerce. Because Apple’s privacy protection inflates this number, always compare against your own history and by email type. A sharp fall from your own average deserves more attention than the gap to any benchmark.
How do you calculate email open rate?
Divide unique opens by the number of emails delivered and multiply by 100. For example, 7,486 opens out of 19,700 delivered emails is 38%. Use delivered rather than sent so you don’t confuse a subject line problem with a delivery problem.
What’s the difference between click rate and CTOR?
Click rate divides clicks by emails delivered and measures engagement across the whole list. CTOR divides clicks by opens and measures how well the content persuaded the people who opened. A campaign can have a low click rate and a high CTOR when the subject line filters the audience well.
What unsubscribe rate is acceptable?
The median in MailerLite’s study was 0.22% per send. More important than any single value is the trend: a steady rise across several campaigns points to excessive frequency or content that isn’t relevant to part of your list.
Why did my open rate go up when I didn’t change anything?
The most common reason is Apple Mail Privacy Protection, which registers automatic opens. It can also be the effect of a list cleanup, since removing inactive contacts raises the average. Check your clicks: if they didn’t rise too, real interest hasn’t changed.
Email is still one of the cheapest channels for selling again to people who already bought. What changed is that the default dashboard in most tools fell behind: it highlights the metric that’s easiest to inflate and hides the one that pays the bills. Teams that start measuring revenue per thousand sends and spam complaints make better decisions with the same data.
The tedious part is reconciling what your email tool says with what GA4 and your CRM recorded. With Sherlok, you connect those sources and ask in plain language, for example "which email campaign generated the most revenue per send this quarter?", and get the analysis ready, no spreadsheet required.
Your AI agents already know what to do in your account.
Turn them on free, no credit card.
- No credit card
- Read-only access
- Revoke it any time