16 June 2026 · 10 min read
B2B Cold Email Reply Rate Benchmarks 2026
Most cold email reply rate numbers you will read in 2026 are vendor showcases, not baselines. The honest figure is lower and falling. Across three large recent datasets, the average B2B cold email reply rate now sits between 3.4% and 5.8%, down from a 6% to 7% band a couple of years ago. This report sets a realistic benchmark, shows what a strong reply rate looks like, and names the levers that actually move it.
Every benchmark below is a vendor or tool study (Tier 3). No neutral institution measures cold email reply rates, so we cite a range across multiple sources rather than treating one figure as gospel. Where a number is old, we say so.
What is a good B2B cold email reply rate in 2026?
A good 2026 reply rate is roughly 5% or higher, against an average of 3.4% to 5.8%. Instantly's Cold Email Benchmark Report 2026 (platform data through Dec 2025) puts the platform average near 3.43%, Saleshandy's 2026 study reports 3.7% across 53.1M emails, and Belkins' 2025 study reports 5.8% across 16.5M emails.
So roughly 4% is the honest median. Beat it, and you are above the field. The wider band exists because the datasets differ: Belkins measures agency-run, heavily targeted campaigns, while Instantly and Saleshandy aggregate huge self-serve platform volume that includes a lot of spray-and-pray.
Across three independent 2025 to 2026 datasets totalling more than 70 million emails, the average B2B cold email reply rate lands between 3.43% (Instantly, 2026) and 5.8% (Belkins, 2025 study). A roughly 4% reply rate is the realistic median; 5% or higher is genuinely good.
| Source | Dataset | Data period | Avg reply rate |
|---|---|---|---|
| Belkins (2025 study) | 16.5M emails, 93 industries | Jan to Dec 2024 | 5.8% (down from 6.8% in 2023) |
| Instantly (2026 report) | Platform data, billions of sends | through Dec 2025 | 3.43% |
| Saleshandy (2026 study) | 53.1M emails | Jan to Jun 2026 | 3.7% |
| Backlinko (historical anchor) | 12M outreach emails | 2019 data | 8.5% (dated, do not use as current) |
Why is the average reply rate falling?
The benchmark is sliding, and that is the real story. Belkins' 2025 study shows the average dropping from 6.8% in 2023 to 5.8% in 2024. Three forces explain it: inbox saturation, the Google and Microsoft bulk-sender rules from late 2024, and Apple Mail Privacy Protection breaking open rate as a usable signal.
For UK and EU teams there is a fourth pressure. Higher send volume raises spam-complaint and consent risk under GDPR and PECR, so the volume tactics that once propped up reply counts now carry legal and deliverability cost. The mechanics of that hygiene live in our deliverability work, so we will not repeat them here.
Stop optimising open rate, start measuring positive reply rate
Open rate is now close to noise. Saleshandy's 2026 study reports a 21% average open rate, while Belkins' 2025 study reports 31% to 32%, after a 2024 peak near 46%. Both are inflated by Apple MPP pixel pre-fetches and bots, so the metric that survives is the positive reply rate: 3% to 5% (Saleshandy, 2026).
Measure the funnel that ends in revenue, not vanity. A clean cold email funnel reads: sent, then delivered, then positive reply, then meeting booked. The positive reply rate, replies showing real interest, is the number to defend in a board deck.
As an Empra operator anchor, not a vendor benchmark, we plan around 2 to 3 booked meetings per 100 well-targeted emails. Treat that as our internal judgment from running these motions, not a measured industry figure.
How does targeting change reply rates?
Targeting is the single biggest lever, and the data is blunt. Belkins' 2025 study shows small, focused sends sharply outperform blasts: campaigns to 50 recipients or fewer reply at 5.8%, while sends of 500 or more collapse to 2.1%. One contact per company replies at 7.8%; ten or more per company falls to 3.8%.
Seniority matters too, though less than people assume. Belkins reports entry-level contacts reply near 8% and C-level near 5%, even as opens run higher for senior titles. The lesson is narrow ICP plus restraint: fewer, better-researched accounts beat scattershot volume. Our outbound work starts from that account list, not a tool.
Belkins' 2025 study of 16.5M emails shows targeting beats volume cleanly. Sends to 50 recipients or fewer reply at 5.8% versus 2.1% for sends of 500-plus. Emailing one contact per company returns 7.8%, while ten or more contacts per company drops to 3.8%. Discipline, not scale, drives replies.
| Variable | Focused | Scattershot |
|---|---|---|
| Campaign size | 5.8% (50 or fewer) | 2.1% (500 or more) |
| Contacts per company | 7.8% (1 contact) | 3.8% (10 or more) |
| Seniority | ~8% (entry-level) | ~5% (C-level) |
Do follow-ups still drive replies?
Yes, and they carry a large share of the pipeline. Instantly's 2026 report finds 42% of replies come from step two onward, and Saleshandy's 2026 study attributes 44% of positive replies to follow-ups. Backlinko's 2019 study (dated, use as a historical anchor) found a single follow-up lifted replies by 65.8%.
But more is not better. Belkins' 2025 study shows diminishing returns and rising spam and unsubscribe rates past the fourth email. The sharp move is sequence depth, not sequence length: three to four well-spaced touches, then stop. Average unsubscribe rates already sit near 2% on cold campaigns (QuickMail via Martal, 2026).
What email copy and length get the most replies?
Short, specific emails win. Belkins' 2025 study reports the highest reply rates at 101 to 200 words, roughly 6 to 8 sentences, at 6.8% to 6.9%, versus about 4% for emails of 600 words or more. Heavy personalisation amplifies this: a Woodpecker study (via Smartlead, 2025) found tailored messages and subject lines lifted reply rates by up to 142%.
The takeaway is plain. Write like an operator sending one message to one person: one relevant trigger, one clear ask, one line of context. Define your terms first if your offer is technical; our GTM glossary covers the shared vocabulary so the email itself stays short.
The elite tier is structural, not lucky
Top campaigns clear a 3x to 5x gap over the average, and they share traits. Instantly's 2026 report puts the top decile at 10.7% or higher. Saleshandy's 2026 study reports the top 5% at 11% to 15% and the top 1% at 15% to 30%. These are not copywriting flukes; they are infrastructure outcomes.
Three structural traits recur: a narrow ICP under roughly 500 accounts, signal or trigger-based sending, and strict per-domain volume caps that protect deliverability. Verified lists matter here too: Saleshandy's 2026 study reports 1.53% bounce on verified lists versus 2.55% on unverified, against Instantly's sub-2% healthy ceiling.
Elite cold email is structural. Instantly's 2026 report puts the top decile above 10.7%, and Saleshandy's 2026 study reports 11% to 15% for the top 5% and 15% to 30% for the top 1%. Those campaigns share a narrow ICP, trigger-based sending, and per-domain volume caps, not a clever subject line.
Multichannel and provider effects
Adding a channel and choosing the right sending platform both move the number. Saleshandy's 2026 study reports email plus a call earns 2.5x the positive reply rate of email alone, and that 97% of cold email is delivered via Google Workspace, ahead of other providers. Channel mix and provider quality compound with everything above.
This is where owned infrastructure earns its keep. Owned sending domains, dedicated IPs, owned and verified data, and AI run as a human-in-the-loop role rather than an unattended bot, all protect the deliverability that every reply rate depends on. Process first, AI second. You can book a working session to map yours.
Key takeaways
- The honest 2026 average B2B cold email reply rate is 3.4% to 5.8% across three datasets totalling 70M-plus emails (Belkins, Instantly, Saleshandy); roughly 4% is median and 5%-plus is good.
- Targeting beats volume: 50-or-fewer sends reply at 5.8% versus 2.1% for 500-plus (Belkins, 2025 study).
- Stop trusting open rate post-Apple MPP; measure positive reply rate (3% to 5%, Saleshandy, 2026).
- Follow-ups drive 42% to 44% of replies, but returns fall and complaints rise past email four.
- The top decile clears 10.7%-plus (Instantly, 2026) through narrow ICP, triggers, and deliverability discipline.
Frequently asked questions
What counts as a positive reply rate versus a total reply rate?
Total reply rate counts every response, including bounces, auto-replies, and "no thanks." Positive reply rate counts only genuine interest. Saleshandy's 2026 study puts the positive reply rate at 3% to 5%. Track positive replies, because they map to meetings and pipeline rather than to inbox noise.
Is open rate still a useful cold email metric in 2026?
Barely. Apple Mail Privacy Protection and bot pre-fetches inflate opens, which is why averages range from 21% (Saleshandy, 2026) to 31% to 32% (Belkins, 2025 study). Use open rate as a loose deliverability hint, never as a success metric. Defend reply rate and meetings booked instead.
How many follow-ups should a cold sequence have?
Three to four touches is the practical ceiling. Follow-ups generate 42% to 44% of replies (Instantly and Saleshandy, 2026), but Belkins' 2025 study shows diminishing returns and rising spam and unsubscribe rates past the fourth email. Add value each step; do not just resend the same ask repeatedly.
Why do small campaigns outperform large blasts?
Relevance and deliverability. Belkins' 2025 study shows 50-or-fewer sends reply at 5.8% versus 2.1% for 500-plus, and one contact per company at 7.8% versus 3.8% for ten-plus. Narrow targeting earns replies and keeps complaint rates low, which protects sender reputation and inbox placement.
Are these benchmark numbers reliable?
Treat them as directional. Every cold email reply rate benchmark cited here is a vendor study (Tier 3), since no neutral institution measures this category. That is why we report a 3.4% to 5.8% range across three independent datasets rather than a single figure, and flag the 2019 Backlinko data as historical only.
About the author
Written by Hugo Dupont, founder of Empra, who builds owned pipeline infrastructure and human-in-the-loop AI roles for B2B teams. Empra's own benchmark, a measured 7.4% reply rate across 1.6M-plus emails sent for 40-plus B2B teams, sits above the 3.4% to 5.8% market average referenced here. Approved Empra proof: £60M-plus pipeline, £8M-plus MRR.
Frequently asked questions
What counts as a positive reply rate versus a total reply rate?
Total reply rate counts every response, including bounces, auto-replies, and rejections. Positive reply rate counts only genuine interest. Saleshandy's 2026 study puts the positive reply rate at 3% to 5%. Track positive replies, because they map to meetings and pipeline rather than to inbox noise and clutter.
Is open rate still a useful cold email metric in 2026?
Barely. Apple Mail Privacy Protection and bot pre-fetches inflate opens, which is why averages range from 21% (Saleshandy, 2026) to 31% to 32% (Belkins, 2025 study). Use open rate as a loose deliverability hint, never as a success metric. Defend reply rate and meetings booked instead.
How many follow-ups should a cold email sequence have?
Three to four touches is the practical ceiling. Follow-ups generate 42% to 44% of replies (Instantly and Saleshandy, 2026), but Belkins' 2025 study shows diminishing returns and rising spam and unsubscribe rates past the fourth email. Add value at each step rather than resending the same ask.
Why do small targeted campaigns outperform large blasts?
Relevance and deliverability. Belkins' 2025 study shows 50-or-fewer sends reply at 5.8% versus 2.1% for 500-plus, and one contact per company at 7.8% versus 3.8% for ten-plus. Narrow targeting earns replies and keeps complaint rates low, which protects sender reputation and inbox placement.
Are these cold email benchmark numbers reliable?
Treat them as directional. Every reply rate benchmark cited here is a vendor study (Tier 3), since no neutral institution measures this category. That is why we report a 3.4% to 5.8% range across three independent datasets rather than a single figure, and flag the 2019 Backlinko data as historical only.