Cold email reply rates in 2026: what benchmarks really say

What is the average cold email reply rate in 2026?

There is no single average. Five benchmarks, read on October 8, 2026, give five different numbers. Every one comes from a company that sells cold email tools or outbound services, and none of the datasets has been independently audited.

Instantly also publishes tiers from the same data: the top quarter of senders reach 5.5% or more, and the top 10% reach 10.7% or more. Hunter's split by use case shows how much the purpose of the email matters: in its data, sales sequences averaged 3%, marketing 6.2%, headhunting 7.5% and digital PR 13%.

Why do cold email benchmarks disagree so much?

Four choices move the number, and each publisher makes them differently.

• What you divide by. Belkins and Instantly divide by emails sent, Saleshandy by emails delivered, and Hunter by people contacted. A three-step sequence sends up to three emails per person, so a per-person rate can be up to three times the per-email rate for the same campaign.
• Which replies count. Instantly counts "all replies received (including follow-up responses)." Belkins takes out auto-replies and bounce notices. lemlist's own formula counts "even an out-of-office message." lemlist's targets and Saleshandy's positive rate count only interested replies.
• Who is sending. Belkins' figures come from one agency's client campaigns. Instantly, Hunter and Saleshandy average across everyone using their tools, from careful teams to bulk senders.
• What the email is for. Hunter's own data runs from 3% for sales to 13% for digital PR. A benchmark that mixes link-building and recruiting outreach with sales email will look better than sales email alone.

Belkins makes the point plainly after changing its own method: "A 5% reply rate against openers and a 0.45% reply rate against total sends can describe the same campaign; they're just measuring different things." Its earlier studies divided replies by people who opened, and it switched to emails sent after turning off its tracking pixel.

How can one campaign have six different reply rates?

Here is a made-up example with round numbers. You email 1,000 people, three emails each: 3,000 sent, 2,940 delivered. Sixty people reply. Twenty of those replies are out-of-office or other auto-replies, 15 say no, and 25 are interested.

Same campaign, same 60 replies, and the headline runs from 0.85% to 6%. Before you compare your numbers with any benchmark, check which line of this table it sits on. In practice a reply usually stops the sequence, so fewer emails go out than in this simple example, which pushes per-email rates up a little.

Where does the 3.43% figure come from?

It is the most quoted cold email reply rate of 2026, and it now has three owners.

• Instantly published 3.43% on January 12, 2026, from its own platform data for 2025.
• Woodpecker's statistics page, updated June 23, 2026, says "The average platform-wide reply rate has declined from 5.1% in 2024 to 3.43% in 2026." The same page's FAQ says the 3.43% is "based on Instantly's benchmark analysis."
• lemlist's blog, updated September 28, 2026, says the figure comes from "Woodpecker's analysis of over 20 million cold emails."

Instantly's report is the earliest dated source we found, and the only one that publishes its own definition for the number. Older figures travel the same way. Snov.io's 2026 statistics page quotes a 5.1% average from Belkins as "recent," while Belkins' current study reports 0.45% on its new method. Woodpecker quotes list-size results from Belkins' earlier 16.5-million-email study (5.8% for lists under 50 recipients, 2.1% for 500 or more), and lemlist repeats them as Woodpecker's data, with "500+" turned into "over 1,000." An older figure, 8.5%, comes from Backlinko's 2019 study of 12 million outreach emails, with data from the outreach tool Pitchbox and outreach that included guest posts and link requests.

When you see a benchmark, look for the publisher's own page, the date of the data, the sample and the definition. If any of the four is missing, treat the number as a rumour.

How many cold email replies turn into meetings?

Meetings are the number that pays, and the published figures disagree even more than reply rates do.

Belkins' figures mean about 3.5% of its replies became appointments, by our arithmetic, far under lemlist's 30% target. Part of the gap is what counts as a reply: Belkins counts every reply that isn't automatic, including people who say no. Part of it is that Saleshandy describes only its best campaigns. Track your own meetings per 1,000 people contacted; it is much harder to flatter than a reply rate.

What do the benchmarks agree on?

The levels disagree, but the direction of most findings holds across datasets. These are vendor figures.

• Smaller lists reply more. Hunter: sequences to 21 to 50 people replied at 6.2%, against 2.4% for 500 or more. Saleshandy: campaigns under 200 prospects got "nearly twice the reply rate" of those over 500. Belkins: "Lower send volumes produce higher reply rates."
• Open tracking costs replies. Hunter: 7.4% without tracking against 4.4% with it. Snov.io, from 44 million emails sent in February and March 2025: 2.36% against 1.08%. Both point the same way, at levels three to four times apart.
• A real domain beats free email. Hunter: 5.2% from a custom domain against 2.5% from freemail addresses.
• Smaller companies and founders answer more. Belkins: 0.72% at companies with up to 10 employees against 0.22% at 10,000 or more; founders and owners 0.57% against 0.32% for VPs.
• A call helps the email. Gong: 3.44% email replies when a rep also called, against 1.81% without, even when the call did not connect.

Follow-ups are where they split. Instantly says 58% of replies come from the first email. Saleshandy says 57% of positive replies come from follow-ups. Hunter found three messages work best, one fewer than in its previous year's data. Read together: the opener does a lot of the work, and two or three follow-ups add a large share on top.

Can you trust a brand-new benchmark report?

Read the method section before the headline. Saleshandy's report, published October 7, 2026 and updated the next day, is a useful example. It gives its study window as "July to October 2026" in one place and "January and June 2026" in another. Its method counts 40,006 sequences, while one table row lists 103,255 sequences for campaigns under 200 prospects alone. Its follow-up table says the first follow-up went to 7.49 million emails at a 0.80% reply rate and produced 29% of all replies. That works out to about 60,000 replies, which implies about 207,000 replies in total, or roughly 0.7% of 30.3 million emails by our arithmetic, against the 4.5% average it reports. The findings may still be useful, but we could not reconcile those figures, so we report them as published and label them.

How should you measure your own cold email reply rate?

1. Report two rates: replies per person contacted and replies per email sent. Say which one you mean every time.
2. Take out auto-replies, out-of-office messages and bounce notices before you count.
3. Count interested replies separately from all replies; only interested replies build pipeline.
4. Track meetings per 1,000 people contacted. It is slow to move and hard to flatter.
5. Compare by use case. Sales email against sales email, recruiting against recruiting.
6. Compare only with a benchmark that uses your method, sample type and year. Write down which one you chose.

Where Duckrow fits

Duckrow's inbox sorts every reply as interested, not interested, out of office or unclear, so interested replies are easy to tell apart from the rest, and you can filter the inbox by campaign and channel. A reply on email, LinkedIn or text ends that lead's sequence. Duckrow does not track opens, and it has no campaign analytics page or meeting booking yet, so meetings are yours to count.