Short answer: A tracking pixel is a tiny image placed in an email. When the message is opened and the image is loaded, your server records an “open”. Today this number is unreliable, because many mail apps block images or load them automatically in advance, and privacy rules treat tracking as personal data. Use clicks, replies and conversions as your main signals, and treat opens only as a rough hint.
How a tracking pixel works
The pixel is an ordinary image, usually one pixel wide and one pixel high, with a unique web address for each recipient. The address looks something like a path on your sending service followed by a long code. When the recipient opens the email and the mail app downloads images, it requests that address from the server. The server sees the request, notes the code, the time and some technical details such as the IP address and the app, and records an open for that person.
Nothing else is needed. No script runs, and nothing is installed. That simplicity is why almost every email platform includes pixels by default, and why open rates became a standard number in reports.
Why open data is now unreliable
Several things distort the numbers, and they work in opposite directions.
- Blocked images. Some mail apps and corporate gateways do not load remote images unless the reader agrees. If the reader never agrees, no open is recorded, even when the message was read.
- Automatic loading by privacy features. Some apps download all images as soon as the message arrives, through a proxy, whether or not the person opens it. This produces opens for messages that nobody read. The article on Apple Mail Privacy Protection explains the best known case.
- Security scanners. Gateways that check links and images for malware may fetch the pixel, creating opens that come from software. See the guide to safe links and URL rewriting.
- Preview panes. A message shown in a preview may count as an open even though the person scrolled past it.
- Text-only reading. Readers who view plain text, or whose app shows only text, are never counted.
The result is a number that can be too high or too low, and you cannot tell which. Using it to decide a subject line test, or to remove “inactive” subscribers, can lead to wrong conclusions.
Privacy and legal points
A pixel collects information about a person’s behaviour, such as when and where they read a message and sometimes the type of device. In many countries this counts as personal data, and in some places it also falls under rules on access to information on a user’s device. For recipients in the European Union, for example, the privacy rules require that you have a lawful basis, that you tell people what you collect and that you respect their rights. The details depend on your location and your audience, so ask a qualified adviser if tracking is central to your business.
A few practical habits help in any case. Say in your privacy policy that your emails may measure opens and clicks. Do not combine open data with other personal profiles beyond what you disclosed. Keep the data only as long as you need it. And never use a pixel in emails where the purpose is sensitive, for instance messages about health or finances, unless you are sure it is allowed and expected.
Does a pixel hurt deliverability?
A single small image is not a spam sign in itself. Mailbox providers see pixels in huge volumes of legitimate messages. Problems appear in other ways:
- Image-only emails. A message that consists of a large picture and little text looks suspicious, as described in the article on image-only emails.
- Tracking domains with a poor reputation. Pixel and link addresses often use the sending platform’s domain. If that domain is shared with senders who behave badly, your mail can suffer. The guide to email links and tracking domains explains how a custom tracking domain helps.
- Broken certificates on the tracking domain. If the pixel address has an invalid certificate, browsers and mail apps show warnings or block it.
- Engagement decisions based on bad data. If you remove or keep subscribers based on opens that are false, your list quality changes in ways you did not intend. See email list hygiene.
Better ways to measure whether email works
- Clicks. A click shows deliberate action. Some are still caused by scanners, so combine click data with later actions on your site.
- Replies. For personal and service emails, a reply is the strongest sign of engagement, and mailbox providers also treat replies as a good signal.
- Conversions. Use campaign links with tags so that your analytics can show sales, sign-ups or downloads that came from a message.
- Unsubscribes and complaints. These show when your content is not wanted. Watch them for each campaign.
- Inbox placement tests. Seed tests and postmaster tools show whether messages reach the inbox, which no pixel can tell you. See Google Postmaster Tools.
When a pixel is still useful
You do not have to remove pixels completely. They can still help in broad comparisons, for example to see whether one send time performs far better than another across thousands of recipients, because errors affect both groups in a similar way. They are less useful for single people. Never trigger an important action, such as a follow-up that claims “we saw you read this”, from a single open event.
If you send transactional email, such as receipts and password resets, you rarely need a pixel at all. The message is expected, and delivery matters more than the open. The article on transactional and marketing email explains the difference.
How to turn pixels off or limit them
Most email platforms have a setting to disable open tracking for a campaign, an account or a domain. If you do this, check that click tracking still works as you expect, and note that your reports will show no open data. Some senders keep tracking on for newsletters and turn it off for service messages. A clear internal rule, written down, avoids confusion when someone asks why the numbers changed.
A short checklist for senders
- Decide per message type whether you need open tracking at all. Receipts, security alerts and replies to support requests rarely do.
- Use a tracking domain that belongs to you, with a valid certificate, so that links and pixels match your sending brand.
- Always include real text in the message. A pixel and one large picture is a weak design for both readers and filters.
- Add alternative text and a plain-text version, so that people whose images are blocked still understand the message.
- Review your reports every quarter and compare opens with clicks. If opens are suspiciously high while clicks and sales are flat, automatic loading is probably inflating the numbers.
What readers see when images are blocked
Many people never see your images at first. They see empty boxes, alternative text or a bar that asks them to load images. A message that makes sense without pictures is therefore better for everyone, and it makes your open data less important. Test your emails with images switched off before every campaign, and put the key message and the main link in text. This also helps accessibility, because screen readers and text-only apps rely on the same content.
How Site AI Audit helps
Site AI Audit does not read your email platform or its reports, so it cannot tell you your open rate. It checks the public setup that decides whether your email is trusted in the first place: SPF with its lookup limit, DKIM signatures, your DMARC policy and your MX records. Good authentication supports delivery whatever you decide about tracking. The free check covers up to 50 pages and needs no sign-up. You can run a free check of your website or see the plans for re-checks and alerts.
Related reading
- Apple Mail Privacy Protection: Why Open Rates Mislead You
- Links in Emails: Tracking Domains, Shorteners and Deliverability
- Email Engagement and Inbox Placement: How Providers Judge You
- Email List Hygiene: How to Clean Your List Without Losing Buyers
The bottom line
Tracking pixels still work technically, but the number they produce has lost much of its meaning. Treat opens as a weak hint, rely on clicks, replies and results, be open about what you collect and keep your sending setup healthy. Good measurement is about decisions you can trust, not about the biggest percentage.
BUJ
Is a tracking pixel legal?
It depends on where your recipients live and what you do with the data. Many places treat it as personal data processing, so tell people in your privacy policy and follow the local rules on consent and rights.
Why do I see opens from people who say they never read the email?
Mail apps and security tools may load images automatically, which records an open without a person reading anything. This is common and cannot be fixed on your side.
Can I block tracking pixels as a reader?
Yes. Turn off automatic image loading in your mail app, or use a privacy feature that hides your IP address and loads images through a proxy. Some apps also warn you when a message tries to track you.
Will removing the pixel improve deliverability?
Not by itself, because one small image is normal. Authentication, list quality, content and complaint rates matter far more.
Should I still report open rates to my team?
You can, but label them as an estimate and put clicks, replies and conversions next to them. That shows the real performance of a campaign.


