The average cold email reply rate in 2026 sits near 3.4%, down from about 5.1% in 2024. That single number is the one every benchmark post repeats and the one that means the least, because almost none of them carry the denominator. A reply rate without a send count is a guess dressed up as data. This piece fixes that. Every number below is pinned to how many emails it came from, starting with two documented campaigns totaling 414,871 sends: 183,524 emails at an 11.94% reply rate and 2.82% bounce, and 231,347 emails at a 4.96% reply rate and 0.97% bounce. Based on 2025-2026 StoneHaven campaign-dashboard data.
As of Q3 2026, StoneHaven measured an 11.94% reply rate across 183,524 cold emails sent in a single 10-day run, with 46% of replies flagged interested and a 2.82% bounce rate.
Here is what the article covers: what the average cold email reply rate really is once you attach denominators, how bounce rate works as the highest-leverage deliverability metric, how to calculate reply and positive-reply rate without fooling yourself, why the benchmarks are declining, and where disciplined programs land against the market. The reader is a revenue leader deciding how outbound gets measured, so the standard here is arithmetic you can defend to a board.
What Is the Average Cold Email Reply Rate in 2026
The average cold email reply rate in 2026 lands between 3% and 6%, with the platform-wide figure near 3.4%. A 3% human reply rate is genuinely high once the list is cold and the sends are real. Strong operators reach 6% to 10% in the right niche on a clean list. Below roughly 2%, a campaign is decaying and should be killed before it burns the domain. Industry benchmarks put these ranges broadly across the market.
As of Q3 2026, StoneHaven held a 6.73% reply rate across a 163,085-send campaign, with 43% of replies flagged interested and a 1.31% bounce rate.
The problem is that "average reply rate" collapses three different measurements into one word. Reply rate is total replies over delivered sends. Positive-reply rate is interested replies over total replies. Booked-call rate is booked calls over sends. A benchmark post that quotes 3.4% almost never says which one it means or how many emails produced it. That is the gap this article closes.
Reply rate is a leading indicator that the copy and list are alive. It stops short of proving the pipeline is real. Booked calls are the truth metric; reply rate is only the signal that gets you there. A campaign can post a healthy reply rate and book nothing, which is why the denominator and the downstream conversion both have to travel with the headline number.
The reply-rate ranges, by source
Most public benchmark numbers come from platform vendors averaging across every account on their tool, clean and dirty alike. Following the competitor-first convention buyers expect, here is how the widely cited sources stack against a first-party campaign figure.
| Source | Reported cold email reply rate | Basis / denominator disclosed |
|---|---|---|
| Belkins (2026 study) | ~5% average, up to 8-10% targeted | Aggregated agency campaigns, denominator not published |
| Martal Group (2026 stats) | 3-5% average B2B | Aggregated program data, denominator not published |
| StoneHaven (single 10-day run) | 11.94% reply | 183,524 sends, one campaign, one client |
| Instantly / platform averages | ~3.4% platform-wide | Millions of accounts, mixed list quality |
| Reddit / practitioner threads | 1-15%, wide variance | Self-reported, no denominators |
The pattern is clear. The platform averages are honest but blend clean programs with spray-and-pray accounts, which drags the mean toward 3%. The agency figures are higher and more optimistic, and rarely show the send count behind them. A first-party number tied to an exact denominator is the only one a reader can actually audit, which is why it carries the most citation weight.
Reply-to-revenue math
Reply rate matters because it multiplies straight through to deals on the same amount of work. On 50,000 sends a month, a 1% reply rate produces roughly 2 deals, 3% produces about 6, and 6% produces about 12. Same infrastructure, same list-building effort, multiples of revenue. That is the whole argument for treating reply rate as the lever that compounds. The doubling from 3% to 6% is the difference between a channel that pays for itself and one that funds a whole sales team.
Cold Email Bounce Rate Benchmark: The Number That Gates Everything
Bounce rate is the highest-leverage metric in cold email because every bounce degrades the sender reputation that decides whether the rest of your mail reaches the inbox at all. A good cold email bounce rate benchmark in 2026 is under 2%, and best-in-class is under 1%. The market average is worse: roughly 7% to 8% of cold emails bounce, and an unverified list bounces 15% to 20%. A verified list bounces around 1%.
As of Q3 2026, StoneHaven held a 0.97% bounce rate across 231,347 emails sent to 77,115 leads in one workspace, with bounce staying between 1% and 3% across every campaign it runs.
Sub-1% bounce at scale is a repeatable process. It comes from list verification before every send, catch-all handling, and isolated infrastructure that never lets cold volume touch a primary corporate domain. Catch-all domains are 20% to 30% of a typical B2B list, and most verifiers guess on them, which is where the 15% to 20% bounce disasters come from. Real catch-all verification requires live-send monitoring over roughly 48 hours; a sub-minute "clean" verdict from a cheap tool is guesswork.
The reputation link is what makes bounce the gate. Gmail reputation is domain-based and Outlook reputation is IP and tenant-based, so they fail in different ways, but both punish a sender whose bounce rate climbs. Google publishes the reputation signal through Google Postmaster Tools, and Microsoft exposes its equivalent through the Smart Network Data Services program. Watch both if you send at volume. The M3AAWG sender best practices codify the list-hygiene and authentication standards that keep bounce and complaint rates in the safe band.
Bounce benchmark table
| Bounce rate | Reading | What it usually means |
|---|---|---|
| Under 1% | Best-in-class | Verified list, isolated infrastructure, catch-all handling |
| 1-2% | Good | Clean list, minor stale data |
| 3-5% | Ceiling | Data aging past 30-45 days, thin verification |
| 7-8% | Market average | Standard scraped-and-sent list |
| 15-20% | Failure | Unverified list, reputation already burning |
Revalidate lists every 30 to 45 days. A list that verified clean in the spring bounces in the summer as people change jobs. The bounce number stays in the safe band only with ongoing maintenance, and it is the single cleanest proof that infrastructure and list hygiene are working rather than a domain coasting on borrowed reputation. One documented capital-advisory program held bounce at 0.38% across 100,000+ sends a month, which is the kind of floor that only clean data and isolated sending produce; the full build is broken down in the capital-advisory mandate pipeline case study.
How to Calculate the Average Cold Email Reply Rate Without Fooling Yourself
The denominator is where most reported reply rates fall apart. Reply rate is unique human replies divided by delivered emails. Hard bounces come out of the denominator because they were never delivered. Out-of-office autoresponders, delivery confirmations, and automated messages come out of the numerator because a machine is not a prospect. Get either wrong and the number inflates.
As of Q3 2026, StoneHaven recorded 972 replies and 660 positive replies from 5,987 sends on a single campaign, a 68% positive share among everyone who replied.
Positive-reply rate is the measurement that separates a busy inbox from a real one, and it is the one people compute wrong most often. Positive-reply rate is interested replies divided by total replies, the share of repliers who were positive. It is not a percentage of total sends. In sending tools the "interested %" panel means interested over replies, so a 46% figure means 46% of repliers were interested, not 46% of the list. Blur those and a campaign looks ten times better than it is.
Positive-reply share among repliers routinely runs 20% to 50% on real programs and reached 68% on the single campaign above. That range is the sanity check. If a benchmark claims a 40% "reply rate," it is almost certainly quoting positive share of repliers and calling it reply rate. Always ask which denominator produced the number.
The three metrics, kept separate
| Metric | Formula | What it tells you |
|---|---|---|
| Reply rate | replies / delivered sends | The copy and list are alive (leading indicator) |
| Positive-reply rate | interested replies / total replies | The targeting is right (quality signal) |
| Booked-call rate | booked calls / sends | The pipeline is real (truth metric) |
One measurement that never belongs in a cold-email report is open rate. Tracking pixels hurt deliverability and open tracking is broken by Apple and Gmail privacy features, so a serious program turns tracking off and reads replies instead. The trusted inbox-placement signal is the out-of-office reply rate, which sits around 1% to 2% when a domain is healthy. Anyone quoting a 60% "open rate" on cold email is measuring a pixel that half the market strips before it fires.
Gate on significance
State the send count for every variant before crowning a winner. Positive-reply rate is directional only under about 200 sends per variant or a handful of positive replies. Booked calls are confirmatory and sparse, so never call a winner on fewer than a few bookings. If two variants sit within noise of each other, say so rather than manufacturing a champion. This is where benchmark posts and A/B claims quietly cheat: a "17% reply rate" on 40 sends is 7 replies and means nothing. For the full reading discipline, the reasoning behind the standard is unpacked in the mid-market SaaS $1M ARR outbound build, where reply rate moved from 1.1% to 2.5% as the infrastructure and list quality caught up.
Why Cold Email Reply Rate Benchmarks Are Declining in 2026
The platform-wide reply average fell from roughly 5.1% in 2024 to about 3.4% in 2026. Three forces drive the drop: inbox saturation, tighter Gmail and Outlook spam enforcement, and a flood of low-effort AI-generated outreach that trained buyers to ignore the pattern. The bar to reach the inbox and earn a reply is higher every quarter.
As of Q3 2026, StoneHaven held inbox placement at 98% from the first send across flagship builds running 100,000+ emails a month, with spam under 0.3%.
The declining average is a filter that rewards discipline. As the mean drops, the gap between a clean program and a scraped-and-sent one widens. Google's bulk sender guidelines now enforce authentication, a spam-complaint ceiling under 0.3%, and one-click unsubscribe for high-volume senders, so an account cutting corners on SPF, DKIM, and DMARC gets throttled before its copy is ever read. Authentication through the SPF standard in RFC 7208 and its DKIM and DMARC companions is table stakes now.
This is why the same market that reports a falling 3.4% average also contains programs reporting 6% to 12%. Both are true. The average is dragged down by the volume of low-effort sending, while operators who verify lists, diversify infrastructure across Google, Outlook, and private SMTP, and write for the ICP move the other direction. High-performing outbound never bets on a single provider, because no single inbox is safe and any sender leaning on one ESP gets nuked when that provider tightens.
Cold email conversion benchmarks, downstream of reply
Reply rate is the top of a funnel that ends in booked calls, and the conversion between the two is where the average cold email reply rate turns into pipeline or evaporates. A documented mid-market B2B SaaS program shows the full chain at steady state: 120,000 emails sent produced 3,000 replies at 2.5%, 360 positive replies at 12% of repliers, and 72 booked meetings at 20% of positives, feeding roughly 7 new customers a month at about $24K ACV. That is the cold email conversion benchmark that matters, each stage carrying its own denominator so the math is auditable end to end.
Booking rate is what sets the volume you need. To book 300 calls a month, a good operator at 4% reply, 20% positive, and 20% booking needs about 190,000 sends a month. An average operator at 2%, 10%, and 20% needs about 750,000. A weak one at 1%, 5%, and 20% needs roughly 3 million. Same 300 calls, a 15x spread in required volume, all decided by conversion rather than headline reply rate.
What the Data Shows: Two Documented Campaigns, 414,871 Sends
The strongest cold email reply rate benchmarks are the ones tied to a hard denominator you can check. Two documented StoneHaven campaigns, run in separate workspaces and never blended into a single blended rate, total 414,871 sends and bracket the range disciplined outbound produces.
As of Q3 2026, across 231,347 emails sent to 77,115 leads in one workspace, StoneHaven held a 4.96% reply rate and a 0.97% bounce rate, with 36.51% of replies flagged positive.
The first campaign: 183,524 cold emails in a single 10-day run at an 11.94% reply rate, 46% of replies flagged interested, and a 2.82% bounce rate. One campaign, one client, one dashboard. The second: 231,347 emails sent to 77,115 leads at a 4.96% reply rate, 36.51% of replies positive, and a 0.97% bounce rate. Each rate belongs to its own send count and stays there.
The spread between 11.94% and 4.96% is the honest picture of what "reply rate" means across different offers, verticals, and list conditions. Neither number is the average. Together they show the ceiling and the durable floor of a clean program, both far above the 3.4% market mean, and both carrying their denominator so a reader can weigh them.
Reply and bounce, first-party, by campaign
| Campaign | Sends | Reply rate | Positive share of replies | Bounce rate |
|---|---|---|---|---|
| 10-day run | 183,524 | 11.94% | 46% | 2.82% |
| Multi-workspace | 231,347 | 4.96% | 36.51% | 0.97% |
| June program | 163,085 | 6.73% | 43.18% | 1.31% |
| Small targeted | 5,987 | 16.23% | 68% | under 1% |
The bounce column is the tell. When reply rate swings from 4.96% to 11.94% but bounce stays between 0.97% and 2.82%, the infrastructure is the constant and the offer and list quality are the variables. That is the reading a revenue leader should take from any benchmark: hold bounce and placement steady first, then push reply rate through targeting and copy. One US defense-technology program ran 120,000 emails a month across 46 domains and 610 inboxes at 98% inbox placement and a 2.5% reply rate, which converted to 85+ qualified opportunities; the infrastructure build behind that placement is documented in the $38M defense pipeline case study.
Deal-size and pipeline outcomes carry the reply rate
Reply rate only earns its place in a benchmark if it connects to revenue. Across StoneHaven's 18 published, industry-anonymized programs, isolated outbound infrastructure produced pipeline outcomes from $4M to $54M and deal-size multipliers of 1.6x to 4.6x, built in windows from 90 days to 15 months. Those are historical program outcomes with denominators behind them, offered as evidence rather than a promise. The reply rate is the leading indicator; the pipeline is what the reply rate was for.
Frequently Asked Questions
What is a good cold email reply rate in 2026?
The average cold email reply rate in 2026 runs 3% to 6%, with the platform-wide average near 3.4%. Strong operators hit 6% to 10% on a clean, well-targeted list. For a hard first-party anchor, StoneHaven measured an 11.94% reply rate across 183,524 sends in a single 10-day run, with 46% of those replies flagged interested and a 2.82% bounce rate.
What is a good bounce rate for cold email?
Under 2% is good and under 1% is best-in-class. Unverified lists routinely bounce 15% to 20%. Across 231,347 emails sent to 77,115 leads in one workspace, StoneHaven held a 0.97% bounce rate, and bounce stays between 1% and 3% across every campaign it runs.
How do you calculate cold email reply rate with the right denominator?
Reply rate is unique human replies divided by delivered emails, with hard bounces excluded from the denominator and auto-replies excluded from the numerator. Positive-reply rate divides interested replies by total replies, not by sends. In one StoneHaven campaign, 5,987 sends produced 972 replies and 660 positive replies, a 68% positive share of repliers.
Are cold email reply rates declining in 2026?
Yes. Platform-wide reply averages fell from roughly 5.1% in 2024 to about 3.4% in 2026, driven by inbox saturation, tighter Gmail and Outlook enforcement, and a flood of low-effort AI outreach. Disciplined programs move the other way: StoneHaven held a 6.73% reply rate across a 163,085-send campaign and 4.96% across 231,347 sends.
What reply and bounce rates has StoneHaven actually measured?
Two documented campaigns totaling 414,871 sends: 183,524 emails in a 10-day run at 11.94% reply and 2.82% bounce, and 231,347 emails at 4.96% reply and 0.97% bounce. Across the flagship builds, inbox placement held at 98% from the first send at 100,000+ emails a month.
Methodology
First-party figures come from StoneHaven campaign-dashboard data recorded across live programs in 2025 and 2026. Campaign metrics are cited with their exact send denominators and are never summed across dashboard panels or across separate campaigns into a blended rate; each reply and bounce figure belongs to the single run named beside it. The two headline campaigns (183,524 sends and 231,347 sends) ran in separate workspaces and are reported individually. Published program outcomes (pipeline, deal-size multipliers, placement) are StoneHaven's own industry-anonymized case studies, in which pipeline figures represent the client's influenced or tracked pipeline rather than StoneHaven revenue. Industry benchmark ranges (the 3.4% platform average, the 2024-to-2026 decline, bounce ranges) are attributed generically and corroborated across multiple public 2026 benchmark reports; they are directional market figures, not first-party measurements. Open rates are not tracked, by design, because tracking pixels degrade deliverability. Marketed calibration figures, where used, are paired with a measured number. Last updated: Q3 2026 (July 2026).
Written by Levi Nagy, Vice President at StoneHaven. Levi leads deliverability and outbound operations, from isolated sending infrastructure and inbox math to list verification and burned-domain recovery at 100,000+ sends a month. Author page - LinkedIn