A good reply rate for a B2B cold email campaign in 2026 is 8% to 15% of emails delivered. Below 5%, something in the list, the copy, or deliverability is broken. Above 15%, you're in the top decile of senders — usually the result of a tightly scoped list and a first line that proves the email wasn't a mail-merge. That range covers every reply, not just the good ones, which is the detail most benchmark posts skip.
What counts as a reply
Reply rate is total replies divided by emails successfully delivered — "interested," "not now," "who is this," and "remove me" all count. That's different from positive reply rate (replies that turn into a meeting or a qualified conversation), which typically runs at 30% to 50% of total replies for a reasonably targeted list. When a tool or a vendor quotes a reply rate without saying which one, assume the bigger, easier number and ask again.
Why the range is so wide
Ready to turn this into pipeline?
Prospect AI runs research, copy, and multi-channel outreach as one system, so consistent pipeline stops depending on heroics.
An 8% list-wide average hides campaigns anywhere from 3% to 20%+, and the spread comes down to two things: how narrow the ICP is and how much the first line proves the sender did homework. A generic list pulled on title and headcount alone, sent a templated email, lands at the bottom of that range even with clean deliverability. A list filtered on a real trigger — a hire, a funding round, a tool in their stack — paired with a first line that references it, routinely clears 15%.
The deliverability floor under every number
None of this matters if the email doesn't land in the inbox. A campaign with strong copy sent from a domain with weak SPF/DKIM/DMARC alignment or a cold IP can show a reply rate near zero simply because most of the send lands in spam and never gets a chance to be read. Reply rate is only a meaningful metric once deliverability is confirmed — check inbox placement before diagnosing copy or targeting.
How to move the number without sending more
The lever with the best return is almost never volume. Narrowing the list to a segment with a real trigger, cutting the first line down to one sentence about the recipient specifically, and following up on a cadence that spreads touches over 2–3 weeks instead of a week (see our sequence-length benchmark) each move reply rate more than adding another thousand contacts. Sending more to the same undifferentiated list just moves the same low rate across a bigger denominator.