Cold email open tracking works by embedding a 1-pixel image in each email. When the image loads, the sender's platform records an open. The assumption is that image load means a human read the email. That assumption has been wrong for years.
How open rate became unreliable
Apple Mail Privacy Protection. In September 2021, Apple introduced MPP for iOS and macOS mail users. MPP prefetches all email content, including tracking pixels, on Apple's servers before the user sees the message. This means every email sent to an Apple Mail user gets counted as opened regardless of whether anyone read it. Apple Mail has roughly 50 to 60 percent market share on mobile. That's a significant chunk of your recipient list generating phantom opens.
Corporate email security scanners. Most enterprise email environments run incoming mail through security filters that click every link and load every image to check for malware. If you're emailing people at corporate domains, their security infrastructure is logging opens before the email reaches the recipient's inbox. The more enterprise your target audience, the worse this problem is.
Combined, these two factors mean a reported open rate on a B2B cold email campaign is almost certainly inflated by 40 to 60 percent with machine-generated activity. A 60% open rate might represent 20 to 30% of actual humans opening.
The curiosity subject line trap
The logical response to "I need to improve my open rate" is to write subject lines that make people curious enough to click. "Quick question," "Thoughts?", "Re: your website," "[First name]."
These do get clicks. They're also annoying to receive when you're the prospect. The reader expected something personal and got a pitch. The annoyance increases the likelihood of a spam report and decreases the likelihood of a positive reply.
Specific subject lines that qualify the reader ("Cold outreach tool for your SDR team?") will have lower apparent open rates because they don't trick anyone. But the people who do open them are genuinely interested. The conversations that come from those opens are more valuable than the conversations from "Quick question?"
Optimizing for open rate leads to copy that attracts unqualified attention. Optimizing for reply rate leads to copy that attracts qualified attention. Those two optimization targets pull in opposite directions.
What to track instead
Reply rate. The only metric that requires a human to read your email and decide to respond. Machines don't reply. A 1.2% reply rate means 1.2% of real humans engaged meaningfully. That number is trustworthy in a way open rate isn't.
Positive reply rate. Total replies include unsubscribes and "remove me" requests. Positive replies, those expressing interest or requesting more information, are the real signal. Track them separately. For most campaigns, positive replies make up 30 to 50 percent of total replies.
Meetings booked per 1,000 sends. The downstream metric that matters most. If you send 1,000 emails and book 4 meetings, your benchmark is 4 meetings per 1,000. Everything else is intermediate.
Pipeline generated per month. Revenue impact. This is what every upstream metric is pointing toward.
When open rate is still worth watching
One context where open rate still provides signal: comparing your own campaigns against each other over time.
If your open rate has run consistently at 45 percent across 20 campaigns and then suddenly drops to 18 percent on a new batch, that's a meaningful change even accounting for tracking noise. It likely indicates a deliverability problem: emails landing in spam, domains getting flagged, or list quality declining. The absolute number tells you nothing, but a sharp relative drop tells you to investigate.
Use open rate as a relative indicator against your own baseline. Not as a benchmark against industry averages, and not as a primary success metric.
The deliverability metric that actually matters
If you want a reliable signal on inbox placement, test it directly. Send a small batch of emails to seed accounts you control at Gmail, Outlook, and corporate domains. Check where they land. Tools like GlockApps and MailReach let you test inbox placement without relying on tracking pixels. That's the real deliverability check, and it doesn't require open rate data at all.
For benchmarks on what reply rates to actually expect, see what is a good cold email reply rate. For the full ROI picture, see Cold Email ROI in 2026.
Frequently Asked Questions
Should I track open rate for cold email?
Not as a primary metric. Apple MPP prefetches email images before humans read them, and corporate security scanners load tracking pixels to check for malware. Together these inflate reported open rates significantly. Reply rate, positive reply rate, and meetings booked per 1,000 sends are the metrics that actually reflect performance.
What is a good cold email reply rate?
For most B2B verticals, 0.5 to 2 percent total reply rate is normal. Above 2 percent is strong. Below 0.3 percent after 300 or more sends suggests a systemic problem with deliverability, offer, or ICP. Positive replies, those expressing interest, typically make up 30 to 50 percent of total replies.
Why does optimizing for open rate hurt cold email results?
Vague, curiosity-based subject lines produce higher apparent open rates but attract unqualified attention. A prospect who opens "Quick question" and finds a sales pitch is more annoyed than one who knew what they were clicking on. Specific subject lines that qualify the reader produce fewer opens but better conversations and higher reply rates from people who are actually interested.
Is open rate ever useful for cold email?
Open rate is still useful as a relative indicator against your own historical baseline. A sudden sharp drop compared to previous campaigns can signal deliverability problems worth investigating. Use it to detect relative changes, not as an absolute benchmark or a primary success metric.