Should you turn off open tracking in cold email?

Open tracking is unreliable and may hurt inbox placement. Track replies instead.

Impact
High
Effort
Low
Updated

Turning off open tracking is a sensible default for cold email. Open counts are unreliable, and the tracking pixel and rewritten links may add deliverability risk, though no independent test of how much turned up in the sources below. Judge your campaign on replies and on a few free signals instead.

Why open tracking is a weak signal

Open tracking places a tiny invisible image in your email. When the recipient's mail app loads it from your tracking server, your tool logs an "open".

Mail apps now load that image without a person reading anything. Apple's Mail Privacy Protection downloads remote content in the background by default, "regardless of whether you engage with the email". Instantly, a cold email tool, adds that Gmail proxies images through its own servers. An open can mean "a machine fetched the pixel", so treat it as a rough hint.

Does the pixel hurt deliverability?

Possibly, and the size of the effect is unknown. Instantly's help center says the pixel "could technically hurt deliverability, but it is difficult to gauge the extent of it". Its blog says some aggressive spam filters may flag emails whose pixels call third-party tracking domains with a poor reputation.

The clearer risk is the shared tracking domain. By default Instantly's tracking links use a domain shared among its customers unless you set up a custom tracking domain, so another sender's spam can rub off on you.

Click tracking adds a similar risk. Instantly's blog says a rewritten tracking link "looks identical to a phishing redirect" and that secure email gateways may quarantine such redirects. Google's sender guidelines separately say web links in the message body "should be visible and easy to understand".

All of this is vendor-reported. The sources below include no independent test that isolates tracking on versus off.

How to do it

  1. Turn off open tracking. In Instantly it sits under a campaign's Options, then Open tracking. Other tools have an equivalent setting.
  2. If you still want click data, set up a custom tracking domain so tracking links use a subdomain you own. Instantly's help center lists the DNS record to add.
  3. Send the first email as text-only if your tool allows it. Instantly's "Send first email as text-only" option sends step 1 as plain text, with no open tracking.
  4. Fix the basics that matter more than any pixel: SPF, DKIM and DMARC records, a warmed sending domain and verified addresses. Google's FAQ lists a missing DMARC record as a reason bulk senders (5,000+ messages a day to Gmail) lose delivery help.
  5. Watch complaints. Google tells senders to keep the user-reported spam rate below 0.1% and never let it reach 0.3%. Gmail's Postmaster Tools shows it.

What to measure instead

Measure replies, and separate human replies from automated ones. Sales.co's data counts every rate against emails sent, not opens, so no open-tracking inflation is involved. Across 1,288,605 cold emails sent from October 2025 to August 2026, 53.7% of replies were automated (auto-responders, out-of-office notices, bounces).

Track bounces too. Instantly's blog says to keep the bounce rate below 2%. For reference, Sales.co's cold campaigns had a 1.71% hard-bounce rate from June 19 to September 22, 2026: 5,047 hard bounces on 294,940 emails.

For website interest, add UTM parameters to the links in your email and read them in your analytics tool. That counts only people who click.

Examples

  • Instantly ships a text-only mode. With "Send emails as text-only (no HTML)" on, opens are not tracked at all, so the tool offers that trade directly.
  • Sales.co reports its own results on sends, not opens, and splits human replies from automated ones. That is the reporting style to copy.

Limits: when it backfires

  • You lose a cheap early warning. If replies collapse, you can no longer see whether mail is landing. Sales.co's data says a raw reply rate below 1% usually points at deliverability or list problems rather than copy, so use that as your alarm and send test emails to your own Gmail and Outlook accounts.
  • Subject-line tests on opens stop working. Test on replies, which needs more sends per variant.
  • It does not repair a bad list. Invalid addresses and missing authentication do more damage than a pixel.
  • Format does not change your legal duties. CAN-SPAM requires a valid physical postal address and a working opt-out, honored within 10 business days, and it makes no exception for business-to-business email.

Questions people ask

Does plain text improve deliverability? HubSpot's tests found HTML and plain-text marketing emails had the same deliverability when the HTML was properly coded and a plain-text version was included. Do not expect a jump from format alone.

Can I keep click tracking? Yes, on a custom tracking domain. Test one campaign each way on your own list.

Sources

Want booked meetings from cold email?

Sales.co runs outbound end-to-end for brands you know: copy, deliverability and replies handled for you.

Talk to Sales.co →

Related tactics

All 133 tactics →