Industry email benchmarks: 5 reasons to compare yourself to yourself

Somewhere in most quarterly reviews, a slide shows up with a borrowed number on it. The industry average open rate is 21.5 percent. Ours is 19.2. The room goes a little quiet, and for the next ten minutes the conversation is about closing a 2.3-point gap to a figure nobody at the table chose, measured on senders we've never met, in a way none of us can actually explain.

That slide feels like accountability. It's usually a detour.

Industry benchmarks are appealing for an honest reason: they look objective. A neutral yardstick handed down from outside the company, immune to the optimism of whoever built the campaign. When you want to know whether you're doing well, an external standard is the most reassuring thing you can reach for.

For one category of email metric, that instinct is exactly right. For the category most teams actually argue about, it quietly leads them astray.

The position of this post, stated plainly: for the numbers that decide whether your email program is working, like opens, clicks, conversions, revenue, and what content earns its place, the benchmark worth trusting is your own history. Not because published benchmarks are fake, but because they answer a different question than the one you're asking. There are two kinds of email metrics, and benchmarks are only honest about one of them.

Two kinds of metrics, two kinds of benchmark

Split your reporting in half.

On one side, hygiene metrics: deliverability, hard and soft bounce rate, spam-complaint rate, unsubscribe rate, list growth and churn. These are governed by rules everyone shares, set by the mailbox providers. Gmail and Yahoo enforce roughly the same thresholds on your account as on your competitor's. A spam-complaint rate of 0.3 percent is a problem whether you sell running shoes or accounting software. Here a universal floor genuinely exists, and an industry benchmark is a meaningful red line.

On the other side, performance metrics: open rate, click rate, click-to-open, conversion rate, revenue per send, and the harder question of which content works for which job. These are governed by things no benchmark can see: your audience, your brand relationship, your offer, how you acquired the list, how old it is, what you sell. There is no universal floor. A "good" open rate for a twelve-year-old retail list and a "good" open rate for a six-month-old B2B list are different numbers, and neither of them is the figure on the slide.

The trouble is that benchmark reports publish both kinds in the same table, in the same font, with the same air of authority. So a 19 percent open rate gets treated the way a 4 percent bounce rate gets treated, as a grade against a standard. One of those is a standard. The other is a coincidence of whoever happened to get surveyed.

Five reasons to compare yourself to yourself

1. An industry average blends senders who share nothing

A published "average open rate for your industry" is a mean across a population with almost nothing in common. It mixes transactional receipts with weekly newsletters, flash-sale blasts with onboarding sequences, a brand mailing two million people with a founder mailing four hundred. The only thing those senders share is a vertical label someone picked on a signup form.

Averaging across distributions that don't overlap doesn't produce a target. It produces a number that sits in the gap between several real ones, describing none of them. We've written before about how comparing a newsletter to a flash sale inside your own account corrupts the comparison. An industry average is that same mistake, scaled up to thousands of strangers.

2. Your audience is the one variable a benchmark can't hold constant

Open, click, and conversion rates are downstream of who's on the list: how they were acquired, how long they've been there, how they feel about your brand, which time zone they read in. Change the audience and every number moves, even when the email is identical.

A benchmark normalizes none of that. It can't, because it has never seen your list. Your own history, by contrast, holds all of it roughly constant. When you compare this quarter's newsletter to last quarter's, the audience is approximately the same audience, so a difference in the numbers points at something you actually changed. That's the entire value of a controlled comparison, and it's available to you for free, against yourself, the moment you stop looking outward.

3. The metrics benchmarks publish are the most contaminated ones

Open rate, the headline of nearly every benchmark report, has been unreliable as an absolute number since 2021, when Apple's Mail Privacy Protection began pre-loading tracking pixels whether or not anyone opened the email. Add security scanners and bots that trip the pixel automatically, and a published "average open rate" is a blend of real human opens and machine noise, in a ratio that differs by audience and shifts over time.

Here's the thing that rescues you: inside your own account, that contamination is roughly constant. About the same share of your list runs Apple Mail this month as last month. So even though your absolute open rate is partly fictional, the delta between two of your own campaigns is still honest, because the noise cancels. Across a benchmark that mixes different provider populations measured in different quarters, it doesn't cancel. It compounds.

4. A benchmark gives you a verdict; your own history gives you a direction

Suppose the slide is right and you're two points under the industry average. Now what? The number doesn't tell you which campaign type is dragging, which content stopped working, or what to change on Tuesday. It's a verdict with no sentence attached, a grade that arrives with no feedback.

Comparing yourself to yourself produces the opposite. "Our newsletter open rate slipped four points over the last six sends, and it started the week we changed the from-name" is a direction. It names the metric, the peer group, the trend, and a suspect. A single A/B test rarely gets you there either; what does is reading patterns across many of your own sends. The industry average can tell you that you're behind. Only your own data can tell you where to go.

5. Only your own data tells you what content works for which job

No benchmark in existence describes your creative. It will never tell you that your single-hero-image layout converts better than your three-column digest, that your editorial tone holds the subscribers your promotional tone burns through, or that your best win-back leads with the offer instead of the apology. Those are the decisions that actually fill a content calendar, and the industry average is silent on every one of them.

Your own history isn't. Every campaign you've sent carries a fingerprint covering layout, density, tone, CTA shape, and subject pattern. Once those descriptions sit next to real outcomes, "what content works for which purpose" becomes a question you can answer from your own archive. Not what works in email in the abstract. What works for you, for this audience, for this job.

When industry benchmarks are actually the right tool

None of this means you throw the benchmark report away. It means you use it for the half of the page it's honest about.

Industry benchmarks are genuinely useful for hygiene. If your bounce rate sits well above the typical range, that's a list-quality problem you can diagnose against an external standard, because list quality follows rules that don't care about your brand. If your spam-complaint rate is creeping toward the 0.3 percent line Gmail and Yahoo enforce, the external threshold is exactly the number to watch. Your own historical average is not the relevant comparison when the mailbox provider is the one keeping score. Same for unsubscribe rate as a deliverability signal, and for the broad question of whether your list is healthy or quietly rotting.

The rule is simple. Where an outside authority sets the floor, meaning the mailbox providers, the spam filters, and the law, benchmark against the industry. Where your audience sets the ceiling, meaning opens, clicks, conversions, and content, benchmark against yourself. Most teams have it backwards: they obsess over the industry's open rate and never look up their own bounce trend.

How comparing yourself to yourself actually works

The reason teams reach for industry benchmarks isn't that they prefer them. It's that comparing yourself to yourself, done by hand, is miserable. It means exporting campaigns from your ESP, whether that's ActiveCampaign, Brevo, or Klaviyo, then sorting them into groups, recomputing averages every time a new send lands, and keeping the whole thing current in a spreadsheet nobody owns. The industry average is just a number you can paste. Effort, not logic, is what usually wins that fight.

So we built the apparatus into Sendlens. It's called clusters: named groups of campaigns that are alike enough that comparing them tells you something real. Newsletter, onboarding, win-back, promotions, announcements. You can sort a campaign into a cluster by hand, or let Sendlens auto-assign it from its fingerprint; when it spots a set of sends that clearly belong together, it will even suggest the cluster for you to create. Each cluster gets its own view and its own running aggregates, including sent, open rate, click-to-open, conversions, and conversion rate, calculated across that group and nobody else.

That's what turns "compare yourself to yourself" from a principle into a default. Your newsletters get benchmarked against your newsletters. Your promos against your promos. "Good" stops meaning "above the figure on the slide" and starts meaning "above our own baseline for this kind of email, and trending the right way." The deeper mechanics of clusters (auto-suggestion, pinned baselines, status) are worth their own read, but the one-line version is this: it's the smallest tool that makes the honest comparison the easy one.

What to do Monday morning

You don't need new software to start. You need to change what the review measures.

1. Split the report in two. Hygiene metrics on one side, performance metrics on the other. Be strict about which is which.
2. Benchmark hygiene outward. Bounce, spam complaints, unsubscribes, and list health: compare those to industry norms and provider thresholds, and act when you cross a line.
3. Benchmark performance inward. For opens, clicks, and conversions, retire the industry-average slide. Replace it with your own trend, grouped by campaign type.
4. Pick four groupings you already argue about. Newsletter, onboarding, promo, win-back is a fine starting set. Report on those, with their own averages, at the next review.

The first time someone asks "is 19 percent good," the answer stops being a shrug at a benchmark and becomes "for our newsletters it's a point above our trailing average, and climbing." That's a sentence a team can do something with.

Frequently asked questions

Are email marketing benchmarks useless?

No. They're useful for one specific job: judging hygiene metrics like bounce rate, spam-complaint rate, and unsubscribe rate, where mailbox providers enforce roughly the same standards on everyone. For performance metrics like open rate, click rate, and conversion rate, your own clustered history is the more honest comparison, because those numbers depend on your audience and offer, which no benchmark can see.

What is a good email open rate?

There is no universal answer, which is the honest answer. A strong open rate for a re-engagement campaign to a dormant list is weak for a welcome email to brand-new subscribers, and Apple's Mail Privacy Protection has made the absolute number unreliable since 2021 in any case. The useful version of the question is "good compared to what," and the most reliable "what" is your own past sends of the same type.

How do I benchmark my email performance against myself?

Group your campaigns into clusters of similar sends, newsletters with newsletters and promos with promos, then compare each new campaign to that group's running average rather than to a flat list or an industry figure. Sendlens does this with clusters that auto-suggest from each campaign's fingerprint, but the principle works in a spreadsheet too: the key move is comparing like with like, over time.

How many campaigns do I need before self-benchmarking is reliable?

Enough that one unusual send doesn't swing the average. As a rough rule, a cluster starts telling you something around five to eight comparable campaigns and gets steadily more trustworthy from there. Below that you are still reading individual results, not a baseline, which is fine, as long as you don't mistake one good Tuesday for a trend.

The comparison that compounds

There's a quiet advantage to grading yourself against yourself that the benchmark misses. Beating the industry average isn't a lever you control; it drifts as other people's lists change. Beating your own best newsletter is a lever you control, and every time you pull it, that send becomes the new floor.

The industry average will never know your name. Your last quarter does.