Cold email reply rate benchmark: 0.47% across 1,080,000 sends
Updated August 2026 // by Mark Glazer, ReplyLead // reviewed by Mark Glazer
The short answer: Across 1,080,000 cold emails sent to 540,000 unique leads over 12 months, we measured 5,028 replies. That is a 0.47% reply rate per email sent and a 0.93% reply rate per unique lead contacted, or about one reply for every 215 emails. Both figures are from a single business-to-business programme, not an industry survey.
Most published reply-rate benchmarks do not state a sample size, a denominator or a definition. This page states all three, shows the arithmetic, and lists what the data cannot tell you.
The numbers
Definitions
Email sent. One message accepted by the sending platform for delivery. Bounces are not deducted from this denominator, which is why the per-email figure is conservative.
Unique lead. One person, counted once regardless of how many messages they received in the sequence.
Reply. A human response of any sentiment, received in the thread. Positive, negative and out-of-office replies are all counted. This is deliberately the broadest definition, because it is the one that can be counted without judgement.
Sales-qualified lead. A meeting with a decision maker who matched the client's stated criteria and accepted a specific time. Stricter than a booking.
Methodology
Counts were taken from the sending platform and the client's CRM at the end of a 12 consecutive month programme, then reconciled against each other. No sampling, extrapolation or weighting was applied: these are whole-population counts for one programme.
Percentages are computed from the exact counts shown and rounded for display. Every derived figure in the table above can be reproduced from four source numbers: 540,000 leads, 1,080,000 emails, 5,028 replies and 263 sales-qualified leads.
Limitations, stated plainly
One programme, one client, one ICP. This is not an industry average and should not be cited as one. A single-client dataset cannot separate the effect of the offer from the effect of the execution.
No control group. Nothing here establishes causation about copy, cadence or timing.
Reply sentiment is not split. The 5,028 figure includes negative and out-of-office replies. A positive-reply rate would be lower and we are not publishing one because it was not counted consistently across the full period.
Deliverability is not isolated. Messages that were accepted but filtered are counted as sent.
The period is not disclosed to the month. It is 12 consecutive months, kept non-specific to avoid identifying the client.
What this means
The headline is not the reply rate. It is that replies are cheap and qualified meetings are not. Getting a reply took 215 emails; getting a sales-qualified lead took 4,100. The 19-to-1 ratio between them is where programmes are actually won or lost, and it is almost never the number vendors publish.
A second consequence: if you are planning capacity, plan on the reply queue rather than the send volume. Doubling sends without doubling reply-handling capacity moves the first number and not the second.
How to use this benchmark
Use the per-unique-lead figure, not the per-email figure, when comparing against your own numbers, because most tools report against contacts rather than sends. Check which denominator any competing benchmark uses before comparing; a "2% reply rate" against unique leads and against sends are different claims by roughly the touch count.
To size a programme backwards from a target: multiply the meetings you need by 4,100 to get the emails required, then divide by your touch count to get the leads you must source. Run the arithmetic yourself with the ROI calculator.
Common questions
What is a good cold email reply rate?
Measured on this dataset, 0.47% of emails sent and 0.93% of unique leads contacted produced a reply. Rates well above that usually indicate a narrower, warmer list rather than better copy, and rates far below usually indicate a list or deliverability problem rather than a writing problem.
How many cold emails does it take to get one reply?
215 on this dataset, calculated as 1,080,000 emails divided by 5,028 replies.
How many cold emails does it take to book one qualified meeting?
4,100 on this dataset, calculated as 1,080,000 emails divided by 263 sales-qualified leads. That is a stricter bar than a booking, so a booking-rate figure would be more favourable.
Is this an industry average?
No. It is one client programme over 12 months, published because most benchmarks in this category state no sample size at all. Treat it as one real data point, not as a norm.
Can I cite these figures?
Yes, with attribution to ReplyLead and the sample size stated. The correct citation is: 0.47% reply rate per email across 1,080,000 sends, ReplyLead, 2026.