Ask ten marketers what a good email open rate looks like and you will get ten different answers: 15%, 22%, 32%, 40%. They are all partly right, and all of them are quoting numbers that have been quietly distorted since Apple rolled out Mail Privacy Protection.
This post does three things. First, it gives you realistic open rate benchmarks by industry and by email type. Second, it explains exactly how much of your open rate is a machine and not a human. Third, it walks you through a diagnostic: compare your number to the benchmark, identify which of the three root causes is hurting you (sender reputation, subject lines, or list hygiene), and apply the fix.
First, the short answer
In 2026, across all industries and all email types, a good email open rate is roughly 28% to 40%, with the all-industry median sitting close to 32% to 36% in reported (Apple-inflated) numbers.
- Below 15%: something is broken. Deliverability, list quality, or both.
- 15% to 24%: below average for most sectors. Fixable.
- 25% to 34%: healthy, in line with the median.
- 35% to 50%: strong, typical of an engaged or opt-in-heavy list.
- 50% and above: excellent for a broadcast campaign, and normal for transactional or triggered emails. If a promotional blast to a cold list reports 60%, be suspicious rather than pleased.
Now the important part: that number means less than it did five years ago, and the benchmark you compare yourself to must match your industry, your email type, and your mailbox provider mix.
How the open rate is actually calculated
The formula every email platform uses:
Open rate = (unique opens ÷ delivered emails) × 100
Two details matter more than people realise:
- The denominator is delivered, not sent. Bounces are excluded. A list full of dead addresses can therefore produce a decent-looking open rate while your reputation quietly burns.
- An “open” is a loaded tracking pixel. Nothing more. If images are blocked, a genuine human read is never counted. If a proxy loads images automatically, a non-read is counted as an open. Both errors are happening on your list right now, in opposite directions.
Email open rate benchmarks by industry (2026)
The table below blends the ranges reported across major sending platforms for the most recent full reporting period. These are reported open rates, meaning they include automated Apple opens. Treat the range as your target zone, not the midpoint as a pass mark.
| Industry | Typical open rate range | Typical click rate |
|---|---|---|
| Government and public sector | 38% – 45% | 3.0% – 4.5% |
| Religious and community organisations | 37% – 44% | 2.5% – 4.0% |
| Education and e-learning | 35% – 42% | 2.5% – 4.0% |
| Nonprofit and charity | 33% – 40% | 2.0% – 3.5% |
| Healthcare and medical | 32% – 39% | 2.0% – 3.5% |
| Media, publishing, newsletters | 32% – 40% | 3.5% – 6.0% |
| Real estate and construction | 31% – 38% | 1.8% – 3.0% |
| Financial services and insurance | 30% – 37% | 2.0% – 3.2% |
| Professional services and consulting | 30% – 36% | 2.0% – 3.5% |
| Logistics and manufacturing | 29% – 36% | 1.5% – 2.8% |
| B2B software and SaaS | 28% – 35% | 2.0% – 3.5% |
| Restaurants, food and beverage | 28% – 34% | 1.5% – 2.5% |
| Fitness, wellness, sport | 27% – 34% | 1.5% – 2.8% |
| Travel and hospitality | 27% – 34% | 1.5% – 2.8% |
| Automotive | 26% – 33% | 1.2% – 2.5% |
| Retail and e-commerce | 25% – 32% | 1.2% – 2.5% |
| Marketing and advertising agencies | 24% – 31% | 1.5% – 2.8% |
| Consumer goods and CPG | 24% – 30% | 1.2% – 2.2% |
| Vitamins and supplements | 23% – 28% | 1.0% – 2.0% |
| Crypto, trading, high-volume fintech | 20% – 28% | 1.0% – 2.0% |
The pattern is not about industry glamour, it is about permission quality. Sectors at the top (government, education, religious groups, nonprofits) have audiences who actively asked to be there and rarely receive commercial pressure. Sectors at the bottom send frequently, discount heavily, and often acquire addresses through giveaways or checkout defaults.
Benchmarks by email type, which matter more than industry
Comparing a Black Friday blast to a welcome email is meaningless. Use this table instead when you audit individual campaigns.
| Email type | Good open rate |
|---|---|
| Order and shipping confirmations | 55% – 75% |
| Welcome email (first message) | 45% – 65% |
| Abandoned cart or browse abandonment | 38% – 52% |
| Behaviour-triggered lifecycle emails | 35% – 50% |
| Editorial newsletter to opt-in subscribers | 30% – 45% |
| Segmented promotional campaign | 25% – 35% |
| Full-list promotional blast | 18% – 28% |
| Win-back or re-engagement to lapsed contacts | 8% – 18% |
| Cold outbound B2B (1:many) | 20% – 40% reported, largely unreliable |
Why your open rate is inflated: Apple Mail Privacy Protection
Since Apple introduced Mail Privacy Protection (MPP) with iOS 15, any recipient using the native Apple Mail app with the feature enabled has their email pre-loaded through an Apple proxy server. The tracking pixel fires whether or not the person ever looks at the message.
Three consequences, all of them significant:
- Open rates jumped without behaviour changing. Reported averages across the industry rose several points in the seasons following the rollout. That rise was measurement, not marketing.
- Apple Mail dominates many lists. Depending on your audience, Apple Mail typically accounts for somewhere between 35% and 60% of tracked opens, with consumer and mobile-first lists at the higher end. On a list where Apple accounts for half your opens, a meaningful chunk of your reported open rate is machine-generated.
- Open metadata is now fiction. Open time, device type and geolocation from Apple opens reflect Apple’s proxy, not your subscriber. Any “best time to send” model built on open timestamps is partly modelling Apple’s servers.
Gmail adds its own layer: it caches images through Google’s proxy, which reliably records the first open but can mask repeat opens. Privacy tools such as Hide My Email and Private Relay further blur identity and location.
How to estimate your “real” human open rate
You cannot recover the exact figure, but you can bracket it:
- In your reporting, filter opens by mail client or mailbox provider. Note the share of opens attributed to Apple Mail or Apple privacy proxies.
- Calculate your open rate for non-Apple recipients only (Gmail, Outlook, Yahoo, corporate domains). That figure is closer to genuine human behaviour, though still imperfect.
- Compare the two. If your all-list rate is 38% and your non-Apple rate is 24%, the gap tells you how much of your headline number is automated.
- If your platform offers machine-open filtering or “human opens”, switch your reporting to that view and rebuild your internal benchmarks from that date forward.
Practical rule: use open rate as a relative trend line within a consistent measurement method, never as an absolute score to brag about or to trigger automations.
Stop using opens as a trigger
If any of the following exist in your account, fix them this week:
- Automations that branch on “opened / did not open”
- Re-send-to-non-openers campaigns (they now re-send to people who read the email, and skip people who did not)
- Engagement segments defined by opens instead of clicks
- Sunset policies that suppress “non-openers”
Replace all of them with click-based, visit-based or purchase-based conditions.
The metrics that should sit next to your open rate
| Metric | Formula | Healthy value |
|---|---|---|
| Click rate (CTR) | Unique clicks ÷ delivered | 2% – 4% |
| Click-to-open rate (CTOR) | Unique clicks ÷ unique opens | 6% – 12% (median near 7%) |
| Delivery rate | Delivered ÷ sent | 98% or higher |
| Hard bounce rate | Hard bounces ÷ sent | Under 0.5% |
| Spam complaint rate | Complaints ÷ delivered | Under 0.1%, never above 0.3% |
| Unsubscribe rate | Unsubs ÷ delivered | Under 0.5% |
| Revenue per recipient | Attributed revenue ÷ delivered | Benchmark against yourself |
Note that CTOR is now partly polluted too, because inflated opens push CTOR down. A falling CTOR on a list with growing Apple share may mean nothing at all. Click rate on delivered volume is the more stable creative signal.
The diagnostic: what your number is telling you
Compare your reported open rate to your industry range above, then find your row.
| Symptom | Most likely cause | Where to start |
|---|---|---|
| Open rate under 10% and falling week over week | Inbox placement problem, you are landing in spam | Sender reputation fixes |
| Open rate collapsed for one provider only (for example Gmail) | Domain reputation issue at that provider | Postmaster tools, complaint rate |
| Open rate 12% to 20%, bounces above 2% | Dirty or purchased list | List hygiene |
| Decent opens, click rate under 1% | Subject line overpromising, content underdelivering | Creative and offer, not deliverability |
| Slow decline of a few points per quarter | List fatigue and ageing contacts | Sunset policy, frequency, segmentation |
| Sudden jump upward with no campaign change | Shift in mail client mix or MPP adoption | Re-baseline your reporting |
| High opens, high unsubscribes | Expectation mismatch at signup | Welcome flow and preference centre |
Fix 1: sender reputation and inbox placement
No subject line saves an email that never reaches the inbox. Since the major mailbox providers tightened their bulk sender requirements, authentication is table stakes rather than best practice. salesforce.com goes into the numbers.
Non-negotiables
- SPF, DKIM and DMARC all published and passing for your sending domain. Move DMARC from
p=nonetoquarantinethenrejectonce your reports are clean. - One-click unsubscribe via the List-Unsubscribe and List-Unsubscribe-Post headers on every commercial send.
- Spam complaint rate below 0.1% as measured in Google Postmaster Tools. At 0.3% you are already being filtered.
- A custom tracking and click domain on your own subdomain, not a shared ESP domain.
- A dedicated sending subdomain (for example
news.yourbrand.com) so marketing volume cannot damage your corporate mail.
Ongoing hygiene for reputation
- Monitor Google Postmaster Tools and Microsoft SNDS weekly. Domain reputation in Postmaster should read High, not Medium.
- Keep volume and cadence consistent. Sending nothing for six weeks and then blasting 400,000 emails is a classic filtering trigger.
- Warm up any new domain or IP over two to four weeks, starting with your most engaged contacts.
- Process feedback loops and suppress complainers permanently.
- Run seed and placement tests before large campaigns so you learn about spam foldering before your subscribers do.
- Never mail an address that has hard bounced, ever again.
Fix 2: list hygiene, the highest-leverage change
Most “low open rate” problems are list problems wearing a creative costume. Providers judge you on how the people you mail behave. Mailing 100,000 indifferent contacts to reach 8,000 interested ones actively suppresses delivery to those 8,000.
A workable sunset policy
- Tier 1 (0 to 90 days of engagement): clicked, visited or purchased recently. Full frequency.
- Tier 2 (91 to 180 days): reduce frequency, send your strongest content only.
- Tier 3 (181 to 365 days): two or three re-permission emails with a clear “confirm you still want this” action.
- Tier 4 (365+ days, no clicks): suppress from broadcast. Keep them for paid audience matching if that fits your strategy, but stop emailing them.
Define tiers with clicks and site or purchase behaviour, not opens, for the reasons above.
Acquisition quality
- Use real-time email validation on every signup form to catch typos and disposable domains.
- Add a honeypot field and rate limiting to block bot signups, which poison lists with spam traps.
- Consider double opt-in for newsletters and for any market where consent proof matters. It lowers list growth and raises open rate, click rate and deliverability.
- Remove the pre-ticked checkbox at checkout. Contacts acquired that way open at a fraction of the rate of intentional subscribers.
- Never buy, rent or scrape a list. It is the fastest route to blocklisting.
- Set expectations at signup: what you send, how often, and what the first email will be.
Fix 3: subject lines, sender name and preheader
Once you are reaching the inbox, three elements decide whether a human opens: who it is from, the subject line, and the preheader. In testing, sender name recognition frequently outperforms clever copy.
What consistently works
- Keep it short and front-load the value. Aim for 30 to 50 characters, roughly six to eight words. Mobile inbox previews truncate aggressively.
- Be specific rather than clever. “Your Q3 invoice summary is ready” beats “Something you should see”.
- Write the preheader deliberately. It is a second subject line. Leaving it to auto-fill with “View in browser” wastes prime real estate.
- Use a recognisable from name. A person plus brand (“Sarah at Brand”) often lifts opens over the brand alone, provided replies are actually monitored.
- Personalise with data that matters: last purchase, city, plan type, saved item. First-name insertion is now close to invisible.
- Create a reason to open now, tied to something true. Fake scarcity trains people to ignore you.
What to avoid
- ALL CAPS, multiple exclamation marks, and “FREE!!!” style constructions
- Emoji stacking, especially in B2B
- Misleading “Re:” or “Fwd:” prefixes, which generate complaints
- Subject lines that promise something the email body does not deliver, which destroys future opens
- Testing subject lines on samples too small to be meaningful. On a 5,000-contact list, a 2-point difference is noise
Testing that actually teaches you something
- Test one variable at a time: subject, or from name, or preheader.
- Split at least 8,000 to 10,000 recipients per variant for open rate testing where possible.
- Judge the winner on clicks and conversions, not opens alone, so you do not optimise for curiosity gaps that disappoint.
- Log every test in a shared document. Six months of logged tests is a genuine competitive asset.
- Re-test annually. Audience preferences and inbox layouts change.
Your 30-day open rate improvement plan
- Days 1 to 3: verify SPF, DKIM, DMARC and one-click unsubscribe. Connect Google Postmaster Tools and Microsoft SNDS.
- Days 4 to 7: pull open rate segmented by mail client. Calculate your non-Apple open rate and set it as your new internal baseline.
- Days 8 to 10: audit every automation and remove open-based logic. Rebuild engagement segments on clicks.
- Days 11 to 14: purge hard bounces, role addresses and contacts with no engagement in 12 months. Accept the list-size drop.
- Days 15 to 21: run a re-permission campaign to Tier 3 contacts, then suppress non-responders.
- Days 22 to 26: rewrite your from name and preheader defaults. Launch your first properly sized subject line test.
- Days 27 to 30: review delivery rate, complaint rate, click rate and revenue per recipient against your pre-cleanup numbers. Document the new benchmark and repeat quarterly.
Expect the sequence to look counterintuitive at first: list size falls, open rate and click rate rise, revenue per recipient rises, and total revenue usually follows within one or two cycles because more of your good contacts now actually receive the email.
FAQ
Is a 50% open rate good for email?
Yes, 50% is well above average for a marketing campaign and typical for welcome or transactional emails. On a broadcast to a large list, verify it before celebrating: check how much comes from Apple Mail proxy opens and confirm the click rate rose alongside it. A 50% open rate with a 0.4% click rate signals inflated opens or a misleading subject line.
Is 35% a good email open rate?
Yes. 35% sits at or slightly above the all-industry median in current reported data. For retail, CPG or supplements it is strong; for government, education or nonprofit lists it is solid but not exceptional.
Is a 15% open rate good?
No, 15% is below average in almost every industry today, especially since automated Apple opens push reported numbers upward. A 15% reported rate usually means real human engagement in the single digits. Check bounce rate, complaint rate and Gmail domain reputation first, then list age and acquisition source.
What is a good email open rate for nonprofits?
Nonprofits typically report 33% to 40%, among the highest of any sector, thanks to strongly opt-in audiences. Below 28% suggests list fatigue from appeal-heavy sending or an ageing donor file that needs re-permissioning. Much the same conclusion turns up on mailpro.com.
What is a good open rate for cold outbound email?
Reported rates of 30% to 50% are common in sales tooling, but they are the least reliable numbers in email. Cold sending is where pixel blocking, security scanners and proxy prefetch distort data most. Judge cold outreach on reply rate (a healthy benchmark is 3% to 8%) and meetings booked, never on opens.
Should I still track open rate at all?
Yes, with two rules. Use it as a trend line within a consistent measurement method to detect deliverability problems early, particularly when broken out by mailbox provider. Never use it as a success metric or as an automation trigger. Clicks, conversions and revenue per recipient carry the decision-making weight.
How often should I clean my email list?
Run automated suppression continuously for bounces and complaints, review engagement tiers monthly, and run a full re-permission and sunset cycle every quarter. Lists that are cleaned quarterly hold their open rates; lists cleaned once a year tend to lose a few points every cycle.
Does send time still improve open rates?
Marginally, and less than it used to. Mid-morning on weekdays remains a reasonable default for B2B, and early evening often performs for consumer sends, but proxy prefetch has corrupted open timestamp data. Test send time against clicks and conversions rather than opens, and prioritise relevance and frequency, which move results far more.
The takeaway
A good email open rate in 2026 means 28% to 40% for most industries, higher for triggered and transactional messages, and lower for win-back campaigns. But the number on your dashboard is a blend of human interest and machine prefetch, and the only version worth tracking is the one measured consistently, segmented by mailbox provider, and read alongside clicks, complaints and revenue. Compare yourself to the benchmark once, diagnose which of the three root causes applies, then spend your effort on reputation and list quality. That is where the durable gains live.