Benchmarks

Average Email Open Rate in 2026: Sourced Benchmarks and What to Track Instead

Published average email open rates range from about 36% to 43%, and every one is inflated by Apple Mail Privacy Protection. Here are the sourced benchmarks, why they disagree, and the metrics worth managing instead.

By Sangeeth · June 2026 · 15 min read

"Is a 35% open rate good?" sounds like a question with a numeric answer. It isn't, because the number in your dashboard is a count of tracking-pixel downloads, and since 2021 a large share of those downloads are made by software rather than people. Before you compare yourself to anyone's average, you need to know what is being counted and how the benchmark you are comparing against was built.

This guide collects published email open rate benchmarks from providers that disclose their methodology, shows where they disagree and why, and then covers what an experienced sender actually does with open data: discount the machine opens, lean on clicks and conversions, and fix the inputs that move real engagement. Every figure below links to the page it came from.

What an email open rate actually measures

An email client cannot tell your sending platform that a person read a message. What the platform can see is a request for a tiny, unique image embedded in the HTML. Amazon SES, for example, describes its open tracking as a 1 pixel by 1 pixel transparent GIF with a unique reference per message: "when the image is downloaded, SES can tell exactly which message was opened and by whom" (Amazon SES metrics FAQ). Most platforms work the same way.

So an "open" really means "something fetched the pixel." A person viewing the message with images enabled triggers it. So does a privacy proxy that downloads images on arrival, and so can a security scanner inspecting the message. A person who reads the message with images blocked, or in a client that clips long messages before the pixel, does not trigger it at all. The SES documentation notes exactly this: when a provider truncates a large message, a pixel placed at the bottom never loads and open rates are understated.

The formula

Platforms calculate open rate against delivered messages, not sent messages, so bounces do not drag the figure down:

open rate = unique opens / delivered x 100
delivered = sent - bounced (hard + soft)

Illustrative example (not benchmark data): you send 10,000 emails, 200 bounce, and 3,920 unique recipients fetch the pixel. Delivered is 9,800, so the open rate is 3,920 / 9,800 = 40%. Some providers report a related figure, open rate against total recipients, which would give 39.2% here. Check which denominator your benchmark source uses before comparing a decimal place.

"Unique" matters too. A recipient who opens five times counts once in unique opens and five times in total opens. Benchmarks almost always use unique opens; raw event logs, including SES's, count every open event separately.

Average email open rate benchmarks (sourced)

The table below compares the headline numbers from three providers that publish their methodology. They are not interchangeable, and the differences are informative.

MailerLite also reports a 2025 click-to-open rate of 6.81%, and notes its 2024 median open rate was 42.35%. GetResponse reports an average click-to-open rate of 8.62%.

By industry

MailerLite publishes medians for 46 industries; Mailchimp's public page lists four plus an all-users figure. Where the categories line up, here they are side by side. Blank cells mean that source does not publish a matching category.

*MailerLite's category is "Higher education"; Mailchimp's is "Education + Training". They are not the same population.

By campaign type and region

The type of email matters as much as the industry. GetResponse reports an average open rate of 40.08% for newsletters, 45.38% for triggered emails, and 83.63% for welcome emails. A welcome email arrives seconds after someone asked for it, so comparing it with a weekly newsletter tells you nothing about either.

Geography matters too. MailerLite's 2025 regional open rates range from 47.69% in Australia and 45.08% in Europe to 31.97% in LATAM, with US and Canada at 44.49%. MailerLite links part of that spread to iPhone ownership: regions where Apple devices are common see more messages automatically marked as opened.

Why the sources disagree

The overall open rates above span roughly eight percentage points, and the industry figures disagree by more. That is expected once you look at how each number is produced:

• Different customer bases. Each provider measures only its own customers. A platform popular with creators and small non-profits will look different from one used heavily by online stores.
• Median versus mean. MailerLite reports medians; Mailchimp and GetResponse report averages. A handful of very small, highly engaged lists can pull a mean up, while a median ignores them.
• Different eligibility filters. Mailchimp counts only campaigns sent to 1,000+ subscribers; GetResponse counts active senders with 500+ contacts. Smaller lists usually open at higher rates, so the cut-off shifts the result.
• Different years. Mailchimp's public data is as of December 2023, GetResponse's covers 2023, and MailerLite's covers December 2024 to November 2025. Apple device share and inbox features change over that period.
• Machine opens. All three pages mention Apple Mail Privacy Protection. None of the headline numbers removes those opens, so every figure above is inflated by an unknown amount.

Apple Mail Privacy Protection and machine opens

Apple shipped Mail Privacy Protection (MPP) with iOS 15 in 2021. Apple's user guide says it "hides your IP address" and "prevents senders from seeing if you've opened the email message they sent you" (Apple: Use Mail Privacy Protection on iPhone). Apple's privacy page describes the mechanism: rather than downloading remote content when you open a message, Protect Mail Activity "downloads remote content in the background by default — regardless of whether you engage with the email," routed through two relays (Apple: Mail Privacy Protection & Privacy).

For a sender, that means every recipient using Apple Mail with the setting on fetches your pixel whether or not they read the message. Those recipients look like 100% openers. The feature is a user setting (Settings > Apps > Mail > Privacy Protection on current iOS), so the share of affected subscribers differs from list to list and you cannot see it directly.

Two consequences follow. First, open rate is inflated by an amount that depends on how many of your subscribers use Apple Mail, which varies by list and region, as MailerLite's regional data suggests. Second, anything downstream of opens degrades: "opened in the last 90 days" segments include people who never read anything, send-time optimisation learns from fetch times instead of reading times, and subject line A/B tests judged on opens can declare the wrong winner.

Apple's guide also states that MPP "does not extend to links." A click still requires a deliberate action, which is why clicks are the more trustworthy engagement signal. They are not perfect either: some corporate security gateways follow links to scan them, which produces clicks seconds after delivery, often on every link in the message.

Other sources of distortion

• Image proxies and caching. The SES FAQ notes that when a provider caches images, only the initial open can be counted, so repeat opens disappear from your totals.
• Images blocked. Real readers with remote images off never register an open.
• Clipped messages. If the pixel sits at the end of a long message and the client cuts it off, the open is lost. SES lets you move the pixel higher for this reason.

Why plain-text emails can't be tracked for opens

Open tracking needs an image, and a plain-text email has no HTML to put one in. The SES documentation is direct about it: "Open tracking only works with HTML emails" (Amazon SES metrics FAQ). A plain-text campaign showing a 0% open rate has not failed; nothing was measuring it.

Clicks can still be measured in plain text if the platform rewrites each visible URL into a redirect link before sending. The trade-off is that the recipient sees the longer tracking URL. In SendDoggie, plain-text campaigns show opens as "Not tracked" rather than a misleading zero, and click tracking stays on unless you switch it off. If you are comparing formats, judge plain-text and HTML sends on clicks, replies and conversions, never on opens. The trade-offs are covered in more depth in plain text vs HTML email.

What to track instead of open rate

Open rate still has two legitimate uses: spotting a sudden deliverability drop (a collapse in opens at one mailbox provider is often the first visible sign of spam-foldering) and confirming a list is broadly alive. For judging content, offers and subscriber engagement, use metrics that need a human action.

The old advice to use CTOR "because it's unaffected by Apple" is backwards: CTOR divides by opens, so MPP inflates the denominator and pushes CTOR down. Click rate against delivered is the cleaner number.

Revenue per recipient is worth adopting even for small stores, because it exposes where email revenue comes from. Klaviyo's 2026 benchmarks report that automated flows produce nearly 41% of email revenue from 5.3% of sends, with click rates of 5.58% against 1.69% for campaigns and revenue per recipient nearly 18 times higher (Klaviyo email benchmarks). A campaign-only open rate would never surface that.

How to segment out machine opens

You cannot remove MPP opens perfectly, but you can stop them from steering decisions. The practical approach is to treat an open on its own as weak evidence and require something stronger before calling a subscriber engaged.

1. Check whether your platform flags proxy opens. Some platforms label opens that arrived through Apple's relay or report a "machine opens excluded" figure. If yours does, use the adjusted number for segmentation.
2. Define engagement by clicks, replies, purchases and logins. A subscriber who clicked in the last 90 days, or bought, or replied, is engaged. One who only "opened" every message within seconds of delivery probably is not.
3. Watch open timing. Background fetches cluster in the minutes after delivery. Humans open over hours and days. A recipient whose opens always land immediately, on every message, at all hours, is a strong MPP candidate.
4. Treat "always opens, never clicks" as unknown, not engaged. Over a long window, a person who reads everything normally clicks something eventually.
5. Discount scanner clicks. Clicks within seconds of delivery that hit every link, including the unsubscribe link, are usually a security gateway, not a reader.

If you can export events, the logic is simple. A sketch of an engagement segment, assuming tables of click and order events:

-- Engaged = a human action in the last 90 days.
-- Opens alone do not qualify.
SELECT c.id
FROM contacts c
WHERE c.status = 'subscribed'
  AND (
    EXISTS (SELECT 1 FROM clicks k
            WHERE k.contact_id = c.id
              AND k.clicked_at > NOW() - INTERVAL '90 days'
              AND k.seconds_after_delivery >= 60)
    OR EXISTS (SELECT 1 FROM orders o
               WHERE o.contact_id = c.id
                 AND o.created_at > NOW() - INTERVAL '90 days')
  );

The 60-second and 90-day thresholds are starting points to tune for your sending frequency, not standards.

What really moves your open rate

Industry explains less of the variance than these four factors, all of which you control or at least know.

List age

Interest is highest just after sign-up, which is why welcome emails open at rates no newsletter reaches. As a list ages, people change jobs, abandon inboxes and lose interest. An old list that has not been pruned will open lower than a young one with the same content, and that tells you about the list rather than the copy.

List source

Subscribers who signed up through your own form, ideally with confirmed opt-in, behave very differently from contacts collected at a trade show, imported from a CRM, or bought. Purchased and scraped lists produce low opens, high bounces and complaints, which damage reputation for everyone else on the list too.

Frequency and consistency

Sending more often usually lowers the per-email open rate while possibly raising total engagement. Erratic sending is worse than either: a list mailed once after six months of silence has forgotten you, and mailbox providers see a sudden volume spike from a sender with no recent history.

Inbox placement

An email in the spam folder is rarely opened. Many "open rate problems" are deliverability problems: missing authentication, poor reputation, or complaint spikes. If opens drop sharply at one provider while others hold steady, look at placement before touching subject lines. See why emails go to spam for the diagnosis.

How to improve your email open rate

These are the changes that move human opens, not just pixel fetches, roughly in order of impact.

1. Authenticate your sending domain

Gmail and Yahoo require SPF, DKIM and DMARC from bulk senders (Gmail sender guidelines, Yahoo sender best practices). A typical minimum DMARC record to start monitoring looks like this:

_dmarc.example.com  TXT  "v=DMARC1; p=none; rua=mailto:dmarc@example.com"

Start at p=none, read the aggregate reports, then move to quarantine once every legitimate source passes. The full setup is in how to improve email deliverability.

2. Use a sender name people recognise

The from name is read before the subject line. Keep it stable and recognisable: your brand, or a person plus brand ("Maya at Acme"). Changing it between campaigns, or sending from a no-reply address, costs recognition you have already built.

3. Write subject lines that state the content

Specific beats clever. "Your March invoice summary" or "3 changes to our API rate limits" tells a reader whether to open; "Quick update!" does not. Use the preheader as a second line that adds information instead of repeating the subject or leaking "View this email in your browser."

4. Clean the list on a schedule

Suppress hard bounces and complaints immediately, and run a re-engagement sequence for subscribers with no clicks or purchases over a long window before removing them. Use clicks rather than opens to decide who is inactive, for the MPP reasons above. The process is in how to clean an email list.

5. Send on a consistent cadence

Pick a frequency you can sustain and keep it. Consistency trains subscribers to expect you and gives mailbox providers a stable pattern to evaluate. If you have been silent for months, warm back up with your most engaged segment first.

6. Segment before you send

A message sent only to people for whom it is relevant earns more opens per recipient and fewer complaints. Recent buyers, trial users and long-term customers need different emails.

7. Test on the right metric

Subject line tests judged on opens are contaminated by MPP, since background fetches happen regardless of subject. Run them on click rate or conversions, with a sample large enough that the difference is not noise.

Common mistakes

• Reporting open rate to stakeholders as engagement. Post-MPP, it overstates attention. Report click rate and revenue per recipient alongside it, or instead of it.
• Comparing against an unsourced benchmark. A number without a dataset, period and definition cannot be compared with anything.
• Mixing campaign types. Blending welcome emails and transactional receipts into newsletter averages flatters the figure.
• Pruning by opens alone. You keep machine-openers who never read anything and may drop real readers who block images.
• Calling a plain-text send a failure because opens show zero. Opens were never measurable in that format.
• Using CTOR as the "MPP-proof" metric. Its denominator is the inflated open count.

Troubleshooting a falling open rate

For deliverability-driven drops, check the raw headers of a test message delivered to the affected provider. You want to see spf=pass, dkim=pass and dmarc=pass in the Authentication-Results header:

Authentication-Results: mx.google.com;
       dkim=pass header.i=@example.com;
       spf=pass smtp.mailfrom=bounce.example.com;
       dmarc=pass (p=NONE) header.from=example.com

Best practices

• Compare open rate against your own trailing average, split by campaign type.
• Cite benchmarks with their source, period and statistic (median or mean).
• Make click rate, conversion rate and revenue per recipient the primary reporting metrics.
• Define "engaged" by clicks, replies and purchases, not opens alone.
• Watch opens by mailbox provider to catch placement problems early.
• Authenticate with SPF, DKIM and DMARC before scaling volume.
• Suppress hard bounces and complaints automatically; re-engage, then remove, long-term inactives.
• Keep sender name and sending cadence consistent.
• Judge plain-text campaigns on clicks and replies.

Key takeaways

• There is no single average email open rate. Recent published figures include 43.46% (MailerLite median, 2025), 39.64% (GetResponse average, 2023 data) and 35.63% (Mailchimp average, data as of December 2023).
• Sources differ because their customers, filters, years and statistics differ. Compare like with like, and always know whether a figure is a median or a mean.
• Every published open rate includes Apple MPP background fetches. Treat opens as a deliverability and trend signal, not proof of reading.
• Plain-text emails cannot be tracked for opens. Measure them on clicks, replies and conversions.
• Click rate against delivered, conversion rate and revenue per recipient are the metrics to manage.
• Real open rate improvements come from authentication, list hygiene, a recognisable sender, specific subject lines and consistent sending.

Sources

1. MailerLite — Email marketing benchmarks (2025 data, December 2024 to November 2025)
2. Mailchimp — Email marketing benchmarks (data as of December 2023)
3. GetResponse — 2024 Email Marketing Benchmarks (2023 data)
4. Klaviyo — Email marketing benchmarks 2026
5. Apple — Use Mail Privacy Protection on iPhone
6. Apple — Mail Privacy Protection & Privacy
7. Amazon SES — Email sending metrics FAQs (open and click tracking)
8. Gmail — Email sender guidelines
9. Yahoo — Sender best practices