45+ Cold Email Statistics You Should Know in 2026

How this list is different

Cold email statistics have a reproduction problem: the same numbers circulate for years, stripped of their datasets, until benchmarks from different denominators get quoted side by side as if they measured the same thing.

Every figure below was traced to a named primary dataset and adversarially fact-checked. Where famous stats failed verification, they are in the debunk section, not the list. And because cold email benchmarks vary wildly by how they are counted, every number here states its dataset and its denominator.

Key cold email statistics at a glance

• Average reply rates range from 0.45% to 3.43% depending on the dataset and denominator (Belkins vs Instantly, 2025)
• The top 25% of campaigns achieve 20%+ reply rates (QuickMail, 65M emails)
• Personalization roughly doubles reply rates across two independent datasets (Belkins; Woodpecker)
• Follow-ups generate 42% to 58.6% of all replies, depending on the dataset
• Founders reply 2-3x more than VPs, and small companies far more than enterprises (Belkins, 7.5M emails)
• Gmail requires bulk senders to keep spam rates below 0.30% (Google, official requirement)

Reply rate benchmarks (read the denominators)

The single most useful thing to know about cold email benchmarks: they disagree because they count differently. Here are the three main datasets, with definitions.

• Average reply rate of 0.45% per email sent, from 34,393 replies across 7,530,489 cold emails, counting unique replies against every email sent, excluding auto-replies and bounces (Belkins, 2025)
• Platform-wide average reply rate of 3.43%, with top performers exceeding 10%, measured per contact across thousands of workspaces (Instantly Benchmark Report, 2025)
• About 50% of campaigns get a reply rate under 10%, while the top 25% of campaigns achieve 20%+ (QuickMail, 65M emails, 2023)

None of these contradict each other: per-email-sent rates sit far below per-contact platform averages, which sit below best-quartile campaign rates. Quote a benchmark without its denominator and you are comparing apples to inboxes.

Open rate benchmarks

• The average cold email open rate is 44% (QuickMail, 65M emails, 2023)
• 8% of campaigns exceed 80% opens; 20% of campaigns sit under 20% opens (QuickMail, 2023)
• Caveat that applies to every open rate anywhere: pixel-based opens are inflated by Apple Mail Privacy Protection, so treat absolute open rates as directional and trust the relative patterns

Personalization statistics

• Personalized subject lines lift reply rates from 3% to 7%, a 133% increase, and open rates from 35% to 46% (Belkins with Reply.io, 5.5M emails, 2024)
• Advanced personalization, custom snippets beyond first name and company, yields roughly 17-18% reply rates versus 7-9% for basic or none (Woodpecker, 20M+ emails; a long-circulated finding from their platform analyses, and a best-subset figure above platform averages)
• Personalized message bodies received 32.7% more replies than non-personalized ones (Backlinko/Pitchbox, 12M emails; flagged: 2019 data, link-building outreach rather than sales)
• The same 2019 study's most sobering number: only 8.5% of outreach emails received any response at all (Backlinko/Pitchbox, 2019, flagged as above)

Follow-up statistics

The three big datasets split on where replies come from, which is itself the finding: sequence design changes the answer.

• 58% of replies come from the first email, so follow-ups contribute 42% (Instantly, 2025)
• The inverse on agency-run campaigns: steps 2-6 generate 58.6% of replies versus 41.4% from the opener (Belkins, 7.5M emails, 2025)
• 55% of replies come from a follow-up step, with 3 follow-ups identified as the optimum (QuickMail, 1.7M-email subset, 2023)
• The consistent takeaway across all three: sequences without follow-up leave roughly half the replies unclaimed, which is why automated persistence matters more than any single email

Subject line and copy statistics

• Question-format subject lines top the field at 46% average opens, versus 44.6% for call-to-action and 44% for number-based lines (Belkins with Reply.io, 5.5M emails, 2024)
• Subject lines of 2-4 words peak at ~46% opens; single-word lines get 38%, and 9-10 word lines drop to 34-35% (Belkins, 2024)
• Elite campaigns keep body copy under 80 words and A/B test weekly (Instantly platform data, 2025)

Send timing statistics

• Morning sends between 8 AM and 12 PM produce the highest reply rate at 0.54%; early morning (5-8 AM) is close behind at 0.52%; late evening and night sends perform worst at 0.40% (Belkins, 7.5M emails, 2025)

Who actually replies

• Founders and owners are the most responsive seniority group at 0.57% reply rate, versus 0.42% for C-level and 0.32% for VPs (Belkins, 2025)
• Reply rate falls almost linearly with company size: from 0.72% at 0-10 employee companies to 0.22% at 10,000+ employee enterprises (Belkins, 2025)
• Scale context: lemlist users alone sent 181,868,356 cold emails in a single year (lemlist lemlab, 2023-24, archived)

Deliverability rules: the numbers that are law

These are not benchmarks but requirements, published by Google for bulk senders (5,000+ messages/day to Gmail), and they define the operating envelope of every campaign above.

• Spam rates in Postmaster Tools must stay below 0.30%, with Google recommending below 0.10% (Google Email Sender Guidelines)
• SPF, DKIM, and DMARC authentication are mandatory for bulk senders (Google)
• Marketing messages must support one-click unsubscribe with a clearly visible link (Google)

Famous cold email statistics, fact-checked

Claims that circulate widely and failed adversarial verification against their supposed sources. They are listed so you can stop citing them.

What the verified numbers actually teach

Strip away the zombie stats and the real playbook is consistent across every independent dataset: write short, personalize beyond the first name, follow up multiple times, send in the morning, aim small and aim high in the org chart, and never let the spam rate approach Google's 0.30% ceiling. Half of all replies live in the follow-ups most senders never send, which is the strongest argument in the entire dataset for automating the persistence rather than relying on memory.

Related benchmarks: our roundups of sales follow-up statistics and B2B cold calling statistics cover the neighboring channels.

Reading the denominators

Almost every disagreement between these datasets comes from one question: reply rate per what.

Per email sent is the lowest number and the most common in vendor reports. Send 10,000 emails across four steps to 2,500 people, get 100 replies, and you report 1%.

Per contact is the number that matters to a sales team. The same campaign is a 4% reply rate, because 2,500 people were contacted, not 10,000.

Per campaign is how the top-quartile figures are usually built, and it skews upward because small, well-targeted campaigns are overrepresented.

None of these is dishonest, but they are not comparable, and a benchmark quoted without its denominator is not a benchmark. Before comparing your own numbers to anything on this page, check which one you are computing. Most CRMs report per contact by default; most sending tools report per email.

One practical consequence: if your reply rate looks unusually bad against a published figure, the first thing to check is not your copy. It is whether you are dividing by the same thing.

The takeaway

Cold email in the verified data: harder than the roundups claim (0.45% per send at agency scale), better than cynics claim (20%+ for the best campaigns), and governed by levers that repeat across every independent dataset, brevity, personalization, persistence, timing, and targeting down-market and up-hierarchy. The famous numbers that failed verification are listed above; everything else here traces to a dataset you can check yourself.

What the numbers mean for your sequence

Three figures do more work than the rest.

Personalization roughly doubling replies, across two independent datasets, is the most reliable finding here. Follow-ups producing 42% to 58.6% of all replies is the second, which makes the first email the smaller half of the job: how many follow-ups to send matters more than most senders assume. The 0.30% spam complaint ceiling is the third, and it is a rule, not a benchmark.

Doing all three by hand is where teams fail. Outsales is an autonomous CRM that writes each message for the person receiving it and decides when the next is due, from your own mailbox. The sending platforms are compared separately.

The sending rules are published directly: Google's sender guidelines define the 0.30% complaint ceiling quoted above.