Cold Email's Strict Reply Rate Is 0.45%, and AI Spam Is Why

The numbers behind the collapse

Belkins measured 7.5 million cold emails sent in 2025 and found a strict reply rate of 0.45%, counting replies against every email sent rather than every email opened. That produced 34,393 replies total. Founders and business owners replied at 0.57%. C-level executives replied at 0.42%. VPs replied at 0.32%. The more senior the target, the worse the rate.

Instantly's 2026 benchmark, measured with a looser methodology across its own platform, puts the average reply rate at 3.43%. Only the top quartile of senders reaches 5.5%. The two figures are not directly comparable since they measure different populations with different methods, but they point the same direction: most cold email gets almost no response.

AI made it worse, not better

Hunter.io analyzed 31 million cold emails for its 2026 State of Cold Email report. The finding that should concern every sales team: 69% of US decision-makers say they are bothered when they can tell an email was written by AI.

That number matters because AI-written outreach is easy to spot. The same personalization patterns show up in every inbox. A line about the prospect's recent LinkedIn post. A reference to their company's latest press release. A compliment that reads like it was generated from a profile scrape. When every email opens the same way, none of them feel personal.

The data backs this up at the campaign level. Emails that were manually edited after AI generation got 18% higher reply rates than unedited AI output, according to the same Hunter.io report. The tool is not the problem. Using it as a replacement for judgment is.

Small lists beat large ones by a factor of 2.5

Hunter.io broke down reply rates by campaign size. The results are not subtle.

Campaigns sent to 21 to 50 recipients averaged a 6.2% reply rate. Campaigns sent to 500 or more recipients averaged 2.4%. That is a 2.5x difference, and it runs directly against how most sales teams use their outreach tools.

The sequence platforms are built for volume. They optimize for sends per day, not for the research that makes a small list work. A team sending 500 emails needs less per-prospect preparation than a team sending 30. The platform economics push toward volume. The reply data says volume is the problem.

The channel preference has already shifted

When Hunter.io asked decision-makers how they prefer to be contacted about business opportunities, 50.5% said LinkedIn. Only 25% said email. The channel that B2B sales was built on is now the second choice of the people it targets.

The top complaint about cold emails was not that they exist. It was that they are too pushy. 65% of recipients said so. The second complaint, at 61%, was irrelevance. Prospects are not rejecting the idea of being contacted. They are rejecting how it is being done.

What this breaks

The SDR model at most B2B companies depends on cold email as the primary outbound channel. At Belkins' 0.45% strict reply rate, an SDR sending 100 emails a day gets fewer than one reply every two days. That reply is not a meeting. It is a response, which may be a request to unsubscribe.

The math is simple. If an SDR costs $70,000 a year fully loaded and generates meetings through a channel that converts at 0.45%, the cost per meeting is higher than most companies' marketing-qualified lead cost through inbound. The channel is being outperformed by the strategy it was supposed to replace.

What to actually do about it

Three sequences of three emails, sent to a list of 30 to 50 people you have actually researched, will outperform 500 AI-generated emails sent to a purchased list. The Hunter.io data says so directly.

Turn off open tracking. Hunter.io found that emails without open tracking pixels got a 7.4% reply rate compared to 4.4% with tracking enabled. The tracking pixel triggers spam filters and signals to the recipient that they are being monitored.

Write the first email yourself. Use AI for research, not for the draft. Edit every email before it sends. The 18% improvement from manual editing is the difference between a channel that works and one that does not.

The decision-makers you are trying to reach have already told you what they want. Shorter emails, sent to smaller lists, written by a person who did the work. The tools that ignore this are the reason the channel is dying.