Click-to-open rate: what it measures and what it cannot

Click-to-open rate is unique clicks divided by unique opens, expressed as a percentage. The idea is sound — it asks "of the people who looked, how many acted?" — and it isolates body and offer performance from subject-line performance in a way that plain click rate does not.

The problem is the denominator. An "open" is recorded when a remote tracking image loads, and since Apple's Mail Privacy Protection began pre-fetching images for Apple Mail users, a meaningful share of recorded opens are machines. Corporate security gateways and some webmail proxies do the same. Clicks are comparatively reliable, since a person generally has to act. So CTOR now has a trustworthy numerator over an inflated, audience-dependent denominator, which means the number is not comparable across audiences and should never be reported as an absolute.

When it is still worth looking at

• As a trend on one stable segment. If the audience composition does not change, the inflation is roughly constant and the direction of travel is informative.
• For body and CTA tests within a campaign. Both arms share the same audience and therefore the same inflation, so a difference between them means something.
• As a cheap anomaly detector. A CTOR that halves overnight usually means something broke — a wrapped link failing, a gateway stripping your URL, images blocked where they were not before.

When it misleads

• Across segments, regions or lists. Apple device share differs by audience, so the denominators differ for reasons unrelated to your email.
• Across time, when your audience is shifting. Growth changes the mix.
• Against anyone else's benchmark. Different tracking setups, different audiences, different inflation. Published CTOR benchmarks are not comparable to yours, and treating one as a target is a good way to make bad decisions.
• For cold outreach. Here it is close to meaningless. Cold email should carry at most one link, often none, and adding a tracking pixel to a one-to-one message adds a remote image and a third-party domain to buy a number you cannot trust.

Raising it legitimately

If you have decided CTOR is the right within-campaign measure, the levers are these, in rough order of effect.

1. Make the subject line and the body promise the same thing. The most common cause of a low CTOR is a subject line that over-promises: you win the open and lose the reader in the first line. Fixing this lowers your open rate and raises your CTOR, which is the right trade.
2. One call to action. Two asks is zero asks. Repeat the same link rather than offering competing ones.
3. Put the CTA in text, above the fold. Not inside an image — images get blocked, and an image-only CTA is invisible to a screen reader.
4. Make the link target match the link text. A mismatch between what the link says and where it goes depresses clicks and raises reports.
5. Keep the email short. Nobody clicks something they did not reach.
6. Segment. The same email sent to a narrower audience gets a better ratio without a word changing.
7. Check rendering, including dark mode and with images blocked. A CTA button that renders as an invisible dark-on-dark block is a common, silent failure.

Link hygiene is a deliverability question too

The way you track clicks has consequences beyond measurement.

• Each redirect domain is a domain whose reputation is not yours. A shared click-tracking domain used by many senders can be filtered or blocklisted for reasons that have nothing to do with you. If you track, use a custom tracking domain on a domain you control, and set it up properly.
• Fewer links is better. Link-heavy mail, especially from a young sending domain, is a weaker signal.
• Never use URL shorteners. They obscure the destination, they are heavily abused, and some receivers treat them accordingly.
• Link text and target should agree. This is both a trust matter and a phishing-pattern matter.
• Security gateways rewrite and sandbox links. Expect some clicks you did not get and some you did, particularly in enterprise audiences.

What not to do

Do not report CTOR as an absolute number to anyone who will compare it to a benchmark.

Do not optimise CTOR by making the subject line vaguer. You can raise the ratio by getting fewer, better-qualified opens — which is genuinely good — or by breaking your tracking, which looks the same in the dashboard.

Do not add tracking to cold one-to-one email for the sake of the metric. Replies are the signal, and the pixel costs you a remote image and a domain.

Do not use CTOR to diagnose a deliverability problem. If placement is bad, your open and click populations are both skewed toward the people who found you in the spam folder.

Do not treat a CTOR drop as a copy problem by default. Check first that links still resolve, that your tracking domain is not blocklisted, and that nothing changed about where your mail is landing.

Where MailPilot fits

Every engagement metric is a ratio over the people who received your mail in a place they could see it, so placement is the variable underneath all of them. MailPilot measures real inbox placement per provider through a seed-inbox network — including whether Gmail is routing you to Promotions rather than the inbox — warms each mailbox over OAuth on Gmail and Outlook or over SMTP/IMAP for SendGrid, Amazon SES, Mailgun, Postmark and custom SMTP, checks SPF, DKIM and DMARC per sending domain, and tracks per-mailbox reputation daily.