Tracking pixels vs link tracking — pick the right signal

A tracking pixel records that an email's images loaded, which may or may not mean a person read it. A tracked link records that someone followed a link, which is a deliberate action and the stronger signal. Use open rate to compare campaigns against each other, and use clicks when a decision depends on real interest, such as whether to follow up.

Most products show both as if they were equivalent signals. They are not. This post explains the difference and how to weigh them.

The tracking pixel

A 1×1 transparent image embedded in the message. When the recipient's mail client renders the email, it fetches the image, and that fetch is recorded as an "open."

Strengths:

• Passive — recipients don't need to do anything for it to fire.
• Cheap — one HTTP GET per recipient per render.
• Familiar — every email tool reports an "open rate."

Weaknesses:

• Many clients block images by default.
• Apple Mail Privacy Protection downloads remote content in the background when a message arrives (Apple), inflating opens.
• Gmail runs a security scan that fetches images seconds after delivery, before anyone opens the message. (Gmail's image proxy, by contrast, fetches when the recipient actually opens it.)
• One human can register many opens (multi-device, image cache misses, re-opens) — so naive counts overstate engagement.

Open rate is best read as a directional signal. Campaign A had a higher open rate than campaign B; some of that delta is real.

Link tracking

Every URL in your email is rewritten to a signed redirect on the tracking host. When the recipient clicks, your tracker receives the request, records who clicked which link, and 302s them to the real destination.

Strengths:

• Active — a click is a deliberate action, much harder to fake than an automated image fetch.
• Per-link granularity — you learn which CTA actually pulled.
• Harder to fake in aggregate — a scanner has to fetch the specific tracked URL, and many announce themselves as prefetches when they do.

Weaknesses:

• Some clients and security products (LinkedIn, Microsoft Defender's Safe Links) scan links in a message for malware, and those fetches can look like clicks. We drop requests that identify themselves as prefetches or previews, but a scanner that doesn't announce itself can still register one.
• A rewritten link is a longer link, and some recipients are spooked by long URLs. We mitigate with a stable tracking subdomain (t.maildriply.com) so the host looks consistent.

How we use both at Maildriply

We surface them separately on the dashboard:

• Open rate is the noisy directional signal. Compare across campaigns; don't take the absolute number to the bank.
• Click rate is the intent signal. If 18% of recipients clicked a CTA, 18% of recipients wanted something.

For follow-up branching in sequences, which are still in development ("send follow-up #2 only if they didn't click"), we weigh clicks over opens. An open alone is too noisy to gate a flow.

TL;DR

• Open rate: directional, noisy, useful for A/B.
• Click rate: high-intent, narrower, useful for branching.
• Reply rate: the truest signal, hardest to observe.

Use all three. Trust them in that order.

For the mechanics behind these numbers, see how Gmail open tracking works and what Apple Mail Privacy Protection does to opens. To set tracking up, start with email tracking in Gmail.