Email opens are tracked by embedding a tiny, invisible 1x1 image (a tracking pixel) in the HTML of your message. When the recipient's mail client loads that image from your server, the request tells you the email was rendered. It is a clever trick, but it has real limits: images are often blocked, and Apple Mail now pre-fetches them automatically, so a recorded open does not always mean a human saw your mail.
How does open tracking work?
When you send an HTML email with open tracking enabled, the sending platform injects a small image tag near the bottom of the body. The image is usually 1 pixel by 1 pixel and fully transparent, so the reader never notices it. The src points back to a unique URL on the tracking server, with an identifier that ties the request to a specific message and recipient.
Nothing happens until the mail client decides to load images. When it does, it makes an HTTP request for that pixel. The server logs the request, records an open event, and returns the image. That event is what shows up in your analytics as an open, often with a timestamp, an approximate location from the IP address, and the user agent of the client.
The mechanics are simple, which is also why they are fragile. The pixel only fires if images load, and image loading is exactly the thing mailbox providers and privacy features have spent years restricting.
Why is open tracking unreliable now?
Three things get in the way, and they have all grown more common over time.
Images are blocked by default in many clients. Plenty of mail apps refuse to load remote images until the reader clicks "show images" or adds you to a contact list. If the images never load, the pixel never fires, and a genuine open goes uncounted. This has always undercounted opens.
Apple Mail Privacy Protection inflates and obscures opens. Since iOS 15 (and the matching macOS release), Apple's Mail Privacy Protection pre-fetches remote content for Apple Mail users, including your tracking pixel, often before the person has even opened the message. It loads the image through Apple's proxy servers, which masks the real IP address and rough location. The effect is twofold: opens get inflated because Apple fetches the pixel whether or not anyone reads the mail, and the location and device data you do get are no longer the recipient's. Apple Mail is a large share of the consumer market, so this is not a rounding error.
Plain-text and preview-only reads slip through. A recipient can read enough in the preview pane to act without ever triggering image loads, and plain-text messages have no pixel at all.
Put together, open rate is now a soft signal. It is useful for spotting large relative changes (a campaign that drops to a third of its usual opens is telling you something), but the absolute number is noisy. Treat it as a directional metric, not a precise count.
What is a stronger signal than opens?
Clicks. A click requires the recipient to take a deliberate action on a link, and click tracking works by rewriting your links to pass through a redirect on the tracking server before forwarding to the real destination. That redirect is hard to trigger by accident, and pre-fetching proxies generally do not follow tracked links the way they fetch images. Click-through rate is a closer proxy for real engagement, and mailbox providers themselves weigh genuine engagement when they decide where your mail lands. If you care about whether people are actually reading and acting on your email, watch clicks and replies more than raw opens. For the broader picture of what to measure, see our email deliverability best practices.
How do you enable or disable open tracking with Bird?
On Bird, track_opens defaults to true, so open tracking is on unless you turn it off. You can disable it per message when you do not want the pixel injected, for example on highly privacy-sensitive transactional mail or plain-text receipts where a stray image tag would be out of place.
The trade-off is straightforward. Leaving it on gives you the open and click data that feeds your dashboards. Turning it off keeps the message lighter and avoids the privacy questions a pixel can raise. Click tracking is configured the same way, and because clicks are the more reliable signal, many senders keep click tracking on even when they are skeptical of open numbers.
You can read which events Bird records and how they are structured in the tracking and metrics guide, and you can explore the resulting reports in email analytics.
FAQ
Does a recorded open mean the person read my email?
Not necessarily. It means the tracking pixel was loaded, which can happen because Apple Mail pre-fetched it, a security scanner opened the message, or a client loaded images in a preview. A real human may or may not have been involved.
Why are my open rates suddenly higher than they used to be?
The most common cause is Apple Mail Privacy Protection pre-fetching your pixel for Apple Mail users. Those automated fetches register as opens and can lift your reported rate well above actual readership.
Should I stop tracking opens entirely?
You do not have to. Opens are still useful for spotting relative changes and for clients that are not affected by pre-fetching. Just pair them with clicks, which are harder to fake and track real intent more closely.
Do open and bounce signals overlap?
No. An open says someone (or something) loaded your message, while a bounce says the message was rejected before delivery. If you are seeing high bounce volume, read what are bounced emails for how those events affect your sender reputation.
Where to go next
Open tracking is worth keeping for the trend lines, as long as you read the number for what it is. Lean on clicks for engagement, watch your delivery and bounce events for reputation, and tune tracking per message when privacy matters. The tracking and metrics guide covers the full event set, and Bird's email product ties sending, tracking, and analytics together.