Cold Email Reply Rates: 2026 Benchmarks by Industry

Every published cold email benchmark measures a different thing, and most teams end up comparing themselves to the wrong one. We run outbound for 50+ B2B companies and sit at a 4.6% reply rate against the 3.43% median Instantly published for 2026. Below, the 5 benchmark studies laid side by side, the splits that actually predict your number, and what each reply rate band tells you about your list.

What Is the Average Cold Email Reply Rate in 2026?

That spread is the first thing to understand, because it is where most benchmarking goes wrong. A founder reads "8.5% is average" somewhere, checks their own 3% campaign, and concludes the copy is broken. The copy is fine. The two numbers were never comparable.

Instantly's 2026 benchmark report, built from campaign data between January 1 and December 18 of 2025, puts the platform-wide average at 3.43%. The distribution behind that average matters more than the average itself. The top quartile starts at 5.5%. The elite tier, meaning the top 10% of all campaigns, clears 10.7%.

Apollo's 2026 guidance lands in the same place from a different dataset: 3 to 5% is a realistic average for a well run campaign, 5 to 8% is strong, and 10% or higher is excellent. Anything under 2% signals a targeting or deliverability problem rather than a copy problem.

The practical read: a 1% reply rate on 10,000 emails produces 100 replies. A 10% reply rate on the same volume produces 1,000. Same sending infrastructure, same tool spend, 10 times the output. Almost all of that gap comes from who you emailed, not what you wrote.

Cold Email Reply Rate

The percentage of cold emails that receive a human response from the recipient. Calculated by dividing total replies by total emails delivered, not by total emails sent. Reply rate is the primary performance indicator for B2B outbound because it measures actual engagement, unlike open rates, which are distorted by privacy proxies and image blocking. A reply includes positive responses, objections, questions and rejections, but excludes auto-replies and out of office messages. In 2026 the working median is 3.43%.

Why Do Published Reply Rate Benchmarks Disagree So Much?

Because reply rate is a fraction, and the industry has never agreed on the bottom half of it. Divide replies by delivered emails and you get one number. Divide by every email sent, including bounces, and you get a much smaller one. Measure a different kind of outreach entirely and you get a much bigger one.

Here are the 5 studies people quote most, side by side, with what each one actually measured.

Belkins is the honest outlier. Dividing 34,393 replies into 7,530,489 sends gives 0.45%, and the firm says plainly that this is a stricter measurement than previous years, which counted against recipients who opened. Nothing is wrong with the number. It is just answering "what does a reply cost me per email I paid to send" rather than "how compelling is my message to someone who received it."

Backlinko sits at the other end for a similar reason. That 8.5% came from link building and PR outreach, where the ask is a link rather than a meeting and the recipient already publishes content for a living. Quoting it as a sales benchmark inflates the bar by roughly 2 times.

So before you compare yourself to anything, write down your denominator. We report against delivered, because bounce is a list hygiene problem with its own fix and folding it into reply rate hides both. If you want the full measurement stack, our guide to tracking cold email campaign performance lays out which metric answers which question.

What Are Cold Email Reply Rates by Industry?

Industry is the strongest single predictor of reply rate after list quality. The same copy and the same infrastructure produce wildly different results depending on who you are emailing, and the driver is inbox saturation rather than how friendly the vertical is.

Source: aggregated from Cleverly's industry benchmark analysis and the Instantly 2026 distribution. Our own industry benchmark breakdown goes deeper on the verticals we run campaigns in.

Worth naming the conflict in the data here. Saleshandy's 53.1 million email sample found SaaS to be the strongest performing industry, while the aggregated picture above and Belkins both put SaaS near the bottom. The difference is who is sending. On a self serve sending platform, SaaS companies are the most sophisticated senders on the tool, so their campaigns beat the platform average. That does not mean a SaaS inbox is an easy target. It means SaaS senders are usually better at this than everyone else.

The strategic implication is simple. If you sell something horizontal, start outbound in the least saturated vertical where you already have proof. You will build revenue faster and generate the case studies that let you go win the harder verticals later. That decision starts with defining your ICP and then segmenting the list so each segment gets its own message.

What Reply Rate Should You Expect by Seniority, Company Size, and List Size?

These 3 splits do more to explain a campaign's number than anything in the copy, and 2 of them are set before a single email is written.

Read the seniority rows carefully, because they look contradictory and are not. Belkins found founders and owners the most responsive title in its 7.5 million email sample, ahead of C-level and well ahead of VPs. Snov.io found C-level replying at 4.2% against 5.6% for everyone below. Both datasets say the same thing in different words: the person who owns the outcome answers, and the more layers of organization sit between that person and the inbox, the worse the number gets. A founder at a 20 person company is a far better target than a VP at a 10,000 person company, even though the VP looks more senior on paper.

Company size confirms it. Belkins measured 0.72% at companies under 10 employees against 0.22% at enterprises, more than a 3 times gap driven entirely by who reads the inbox.

List size is the loudest effect in the whole dataset, and it is the one people misread. Saleshandy found campaigns under 200 prospects reply at 15 to 20% while campaigns of 500 to 1,000 reply at 8%. Woodpecker found the same shape at smaller scale: 5.8% under 50 contacts, 2.1% above 500. Small lists are not magic. They are a proxy for how much research went into each row. A 150 person list is one somebody actually built. A 5,000 person list is usually one somebody filtered.

That does not mean you cannot send volume. It means volume has to be assembled from many small, tightly defined segments rather than one wide pull. Our guidance on how many emails to send and on scaling past 1,000 emails covers how to hold precision while the number goes up.

What Is a Good Positive Reply Rate?

Raw reply rate counts every response, including "remove me" and "not interested." A 6% reply rate where 90% of the replies are rejections is worse than a 3% reply rate where half are interested. Positive reply rate is the number that actually tracks with booked conversations.

Positive Reply Rate

The percentage of delivered cold emails that produce a response showing genuine interest: a meeting request, a qualification question, a request for more detail, or a clear expression of intent. It excludes opt outs, rejections, referrals away, auto-replies and out of office messages. Positive reply rate predicts booked conversations far better than raw reply rate, which is why it is the number outbound teams should report to the business.

In strong campaigns, interested replies make up roughly 25 to 50% of total reply volume, per Saleshandy's 53.1 million email sample. Apollo's working bands for an SDR team are 2 to 3% positive reply rate as minimum viable, 4 to 6% as the target, and 8% or higher as the stretch.

That third number is the one most teams sleep on. Saleshandy measured soft calls to action producing 78% more positive replies than hard ones, and campaigns running multiple competing asks performing worst of all. One ask, low friction, is the pattern. We go deeper on that in writing a cold email CTA and on where the hook ends and the ask begins.

To report positive reply rate at all you need a consistent classification layer, because "interested" is a judgment call that has to be made the same way every time. Our approach is in reply classification and the full definition sits in what is a positive reply rate. Even the rejections carry information, which is why handling not interested replies is a worked process rather than a delete key.

What Does Each Reply Rate Band Tell You About Your List?

This is the part the benchmark posts skip. A reply rate is a diagnostic reading, and each band points at a different broken part. Fixing in the wrong order costs weeks.

Under 0.5%: check placement before you rewrite anything

A campaign sitting near the Belkins per-send average while measured against delivered is not a messaging problem. It is a delivery problem, and no amount of copy work moves it. Run an inbox placement test, read domain reputation, and confirm authentication is actually passing before anything else. Google requires bulk senders to keep the spam rate reported in Postmaster Tools below 0.3%, and recommends staying under 0.10%. Crossing that line quietly turns a working campaign into a silent one.

0.5% to 2%: the list, not the copy

This is where most teams jump straight to rewriting emails, and it is almost always the wrong move. Under 2%, the usual cause is that you are emailing people who were never going to buy. Tighten the ICP until you can name the trigger that makes someone a buyer this quarter, then rebuild. Our take on building a list from scratch covers the sourcing order.

2% to 3.5%: right people, forgettable message

The targeting is fine and the message could have been sent to anyone. Backlinko measured a 32.7% lift in response from a personalized body and 30.5% from a personalized subject line across 12 million emails. Woodpecker put advanced personalization near 17 to 18% reply rate against 7 to 9% for basic. The bar has moved: first name and company name are table stakes now. See personalizing at scale for how to hold that standard across thousands of sends.

3.5% to 5.5%: the sequence is too short

Instantly found the first email produces 58% of replies and follow-ups produce the other 42%. Saleshandy, looking specifically at positive replies, found follow-ups produce 44% of them, with the first follow-up alone contributing 26%. Backlinko measured a 65.8% lift from adding one follow-up. If you are running 2 steps, you are collecting a little over half of what the list will give you. The follow-up sequence guide has the cadence, and when to stop following up has the other end of it.

Above 5.5%: read the mix, not the headline

At this level the question stops being "how many replied" and becomes "how many of them were real." Break the number into positive, objection, not now and hard no. A campaign at 7% with a 20% positive share is weaker than a campaign at 4% with a 50% positive share, and only one of those two is worth scaling.

What Actually Moves a Reply Rate, in Order?

After 50+ B2B campaigns, the list of variables that genuinely move the number is shorter than most people expect. Here they are in the order we work them, because doing them out of order wastes the most time.

1. Inbox placement. Everything downstream multiplies by this. Get deliverability confirmed first: SPF, DKIM and DMARC passing, DNS records clean, warmup complete, and spam placement ruled out. A 2% reply rate with 40% of mail in spam is a 3.3% campaign wearing a disguise.
2. List precision. The largest measured effect in the data. Cut the bottom 30% of the list, the rows that technically match the filter but have never bought from anyone like you. Then build the rest from segments rather than one wide pull.
3. Verification. Saleshandy measured verified lists bouncing at 1.53% against 2.55% unverified, roughly 40% less bouncing. Every bounce makes the next 100 emails slightly less likely to land. Verification is the cheapest insurance in outbound, and bounce rate causes and fixes covers what to do when it is already high.
4. Message specificity. One verifiable detail per email that could not apply to any other company on the list. That is the whole standard. Writing cold emails that get replies has the structure.
5. Length and ask. Elite performers average under 80 words on the first touch, per Instantly, and Saleshandy puts the best subject lines at 4 to 5 words. Soft ask beats hard ask by 78% on positive replies. Brevity reads as confidence. Length reads as need.
6. Sequence depth. Four to 7 steps, each adding something new rather than repeating the ask. Timing matters less than most think: Instantly puts peak engagement on Wednesday, Belkins on Wednesday and Thursday, and send timing is worth about one tenth of what list quality is worth.

One more from the data that surprises people: Snov.io found campaigns with open tracking disabled replying at 2.36% against 1.08% with tracking enabled. The tracking pixel is a deliverability tax. Given that open rate is already unreliable, that is a trade most teams should take, and it is a big part of why we treat open rate as a diagnostic rather than a KPI.

Infrastructure carries more of the number than copy does, which is why sending is spread across multiple domains on secondary domains, each one warmed properly, with monitoring running the whole time. The setup order is in setting up email domains for outbound.

How Do You Turn Reply Rate Into Booked Conversations?

Reply rate is a leading indicator. The business cares about meetings, so the honest math runs from sends to conversations.

Saleshandy's number for a well run sequence is 2 to 3 meetings per 100 emails, which is the upper end of the SMB row and well above what high ticket campaigns produce. That gap is the deal size trade. For a high ticket B2B offer, 3 to 8 meetings per 1,000 emails sounds thin until you run the math on a 5K per month deal at a 25% close rate. Five meetings from 1,000 emails produces 1 to 2 clients worth 60K to 120K a year. The economics work because the deal justifies the volume, and ROI by ACV shows where that stops being true.

Where it gets interesting is what happens when you change the ask. Our reply rate across 50+ campaigns sits at 4.6% against the 3.43% market median, and the single biggest contributor is not copy quality. It is that the ask on a podcast invite is a compliment rather than a sell. Nobody guards their calendar against being interviewed. The invite reply rate benchmarks break that out, and invite versus pitch is the strategic case for it.

That is also why we report the unit as a recorded conversation rather than a meeting. It is a specific thing: an ICP decision maker who shows up and completes the recorded interview. Our commitment is 30 of those in 90 days or your money back, editing included, invites by email only. How many invites it takes to book one recording has the working ratio, and what counts as a qualified meeting covers the definition problem underneath all of this.

Why Reply Rate Is Still the Metric We Run On

Open rate is unreliable and, per Snov.io, tracking it costs you replies. Click rate depends on having a link in the email, which most cold emails should not have. Bounce rate is hygiene, not performance. Reply rate is the only number that confirms a real person read the message and cared enough to answer.

The right way to hold it is as a pair. Report reply rate for volume health and positive reply rate for quality, and never let one travel without the other. A number moving up while the positive share falls means the message got louder, not better.

The strongest outbound operations in 2026 are not the ones sending the most email. They are the ones producing the highest ratio of positive replies per email sent, which improves by sending fewer and better emails to a more precisely defined list. Volume is the easy lever. Precision is the one that keeps working at scale, and the ask you choose is the one nobody is testing.