Email Open Rate: What Cold Senders Can Trust

How is email open rate calculated?

The standard formula is unique opens divided by delivered emails, multiplied by 100. Delivered means sent minus bounced. Leaving bounces out of the denominator matters: if you divide by everything you sent, a dirty list drags the rate down and you end up blaming the subject line for what's really a data problem. Unique opens count each recipient once, however often they reopen the message. Total opens count every fetch, so one prospect forwarding your email to three colleagues can push a single row up to four. Most sending platforms show both, and they aren't interchangeable. Compare campaigns using the same one. Also keep the time window fixed. Opens keep arriving for days, so a rate pulled two hours after sending and one pulled a week later describe different things. One last point that people miss: the formula can only count opens it can detect. Every limitation in the sections below comes from the detection step, not from the arithmetic.

How does an email open actually get recorded?

An open is recorded when the recipient's mail client requests a tiny image, usually 1x1 pixel, with a URL unique to that recipient. The server logs the request and marks the email as opened. No image request, no open. That's the whole mechanism, and it's why the number fails in predictable ways:

• Images blocked by default: classic Outlook desktop doesn't download external images until the user allows it, so a real reader may never be counted.
• Plain-text emails: a text/plain message can't embed an image, so it can't be tracked for opens at all.
• Proxy prefetching: Apple Mail Privacy Protection, introduced with iOS 15 and macOS Monterey in September 2021, loads remote content through Apple's proxies whether or not the person opens the message.
• Image caching: Gmail has served images through its own proxy since December 2013. That hides the reader's IP address and location, and repeat opens may never reach the tracker.
• Security scanners: corporate mail gateways often fetch every image and link to check for threats before delivery, which records an open no human made.

B2B inboxes are where the last item hits hardest, because that's where gateway scanning is common.

Why is my email open rate suddenly so high?

An open rate that jumps without any change on your side almost always comes from machines fetching the tracking pixel. Apple Mail Privacy Protection is the usual culprit. If a meaningful share of your prospects read mail in Apple Mail with the feature turned on, every one of those messages registers as opened soon after delivery. A new corporate security gateway at a target account does the same thing, and it often clicks links too. The opposite happens as well. A sudden drop can mean a recipient domain started blocking images, or your mail started landing in spam, where most clients don't load images. Since a spam-folder placement and a real lack of interest look the same in the open column, a falling open rate is a reason to check authentication and placement, not proof that your copy got worse. Check that SPF, DKIM and DMARC still pass for your sending domain before rewriting anything.

Cause	Effect on open rate	Why
Apple Mail Privacy Protection	Inflated	Proxy prefetches the pixel without a human opening
Corporate security scanner	Inflated	Gateway fetches images and links before delivery
Outlook image blocking	Deflated	Pixel not requested until the user allows images
Plain-text email	Not measurable	No image can be embedded
Spam-folder placement	Deflated	Most clients don't load images in spam

What does an email example with an open-rate calculation look like?

Take a cold email to an operations lead: a subject line naming their company's new warehouse, two sentences saying why you're writing, one question, and a sign-off with an unsubscribe line. If it's HTML, the tracking pixel sits invisibly at the bottom. Now the numbers, which are illustrative. You send 400 of these. 12 hard-bounce, so 388 are delivered. Your platform shows 150 unique opens, so the open rate is 150 ÷ 388 = 38.7%. Now the part the dashboard doesn't show. Suppose 60 of those opens came from Apple's proxies a few seconds after delivery. Those 60 tell you the email arrived, not that anyone read it. Meanwhile, some Outlook users who read every word never show up at all. The 12 bounces are the hard data here: 3.0% of what you sent went to addresses that don't exist. That's a list-quality problem you can fix today. For copy-ready text to start from, see our sample cold email examples.

What should cold email senders track instead of open rate?

Track the numbers that a receiving server or a human produces directly, not ones inferred from an image request. Mailbox providers base sender reputation on signals like these, so they also tell you more about whether your domain is healthy:

1. 1 Hard bounce rate: addresses that don't exist. Clean the list before it costs you reputation.
2. 2 Spam complaint rate: Gmail's bulk sender rules from February 2024 ask senders to stay below 0.3% as reported in Postmaster Tools.
3. 3 Replies: a human typed something back. Count positive and negative separately.
4. 4 Unsubscribes: for senders of 5,000 or more messages a day to Gmail, one-click unsubscribe under RFC 8058 is required.
5. 5 Authentication results: SPF, DKIM and DMARC pass or fail, visible in the message headers and in DMARC aggregate reports.

Keep watching open rate, but read it as a trend within one list over time, not as a score. If a new domain shows weak numbers everywhere, its warm-up is the first thing to look at. Our email warm up guide covers daily volumes per week.

Does sending video change how email open rate is measured?

No. Opens still depend on a remote image being fetched. Any preview image loaded from a server counts as remote content, so Apple's proxies prefetch it and Outlook may block it, just like a pixel. What EmailFlow OS changes is how the video gets made and where it's sent from. You record one video and upload it to EmailFlow OS. Then you connect your Brevo account and sending domain and import your prospect list into EmailFlow OS. EmailFlow OS makes a personalised version of that one recording for each recipient, not one per company, and sends each version from your own inbox through your own Brevo account and your own domain. It never goes out through a shared platform sender. That means the reputation built by those sends belongs to your domain, and the sending history sits in your Brevo account where you can read it alongside the bounce and complaint data above. How a version per recipient works is explained in our article on personalized video messages.

Questions people ask

? What is a good email open rate for cold email?

There's no reliable benchmark. Since September 2021, Apple Mail Privacy Protection has recorded opens that no human made, and Outlook's image blocking hides real ones. So two senders with the same 40% may have very different numbers of actual readers. Compare against your own earlier sends to the same kind of list, and judge results by replies and bounces.

? How do you calculate email open rate?

Divide unique opens by delivered emails and multiply by 100. Delivered means sent minus bounced. Example: 150 unique opens out of 400 sent with 12 bounces gives 150 ÷ 388 = 38.7%. If you divided by all 400, you'd get 37.5%, and the gap would grow as your list quality gets worse.

? Can you track opens on a plain-text email?

No. Open tracking works by having the client load a remote image, usually a 1x1 pixel. A text/plain message can't contain an image, so there's nothing for the client to load and no open gets recorded. Replies, bounces and unsubscribes still work as signals, because they don't depend on images.

? Does Apple Mail Privacy Protection count as an open?

Yes, in most tracking systems. Apple Mail with Privacy Protection enabled downloads remote content, including tracking pixels, through Apple's proxy servers, often soon after delivery and whether or not the message is read. The tracker sees a request and logs an open, but the request tells you only that the message reached an Apple Mail inbox.