Cold Email Statistics 2026: 40 Numbers From Real Campaigns, Not Surveys

This page is a reference list of cold email statistics for 2026, measured inside the live client campaigns we run at Pipeline Playbooks, an agency sending up to 50,000 cold emails a month per client. These are operating numbers, not survey answers: the average reply rate at volume is 0.5 to 0.6 percent, 10 percent or more of replies being interested marks a good campaign, and an offer is proven dead at roughly 2,000 sends and around 20 replies with zero interested.

Most cold email statistics online are surveys quoting other surveys. Every number below traces to a full post on this site where we show the working, and each statistic links to its source. Use this page as the index and the linked posts as the evidence.

How were these cold email statistics measured?

Every statistic on this page was measured inside live client campaigns. The numbers below are the doctrine those campaigns actually run on: the thresholds our monitoring is wired to and the benchmarks we grade every campaign against.

Three measurement rules keep the numbers honest. Auto-replies, out-of-office messages, and unsubscribes are excluded from reply stats on every campaign. Every interested classification is verified by hand against the actual reply text before it counts, because automated classifiers over-report: in one weekly review a dashboard claimed 9 interested replies and the verified number was about 4. And no verdict is issued below the volume floors in the testing section, because a bad number on thin volume is noise, not evidence.

We earned that discipline the expensive way. One of our own campaigns ran to 13,000 sends and 28 replies with zero interested before anyone summed the numbers at the offer level, because every per-variant row looked too thin to call. The full autopsy is in anatomy of a dead cold email campaign, and it is why this page exists: the numbers only protect you if you read them at the right level.

What are the key cold email reply rate statistics?

• 0.5 to 1.0 percent is a healthy cold email reply rate at volume, 1.0 to 1.5 percent is good, and above 1.5 percent is really good. (Cold email reply rate benchmarks)
• 0.5 to 0.6 percent is the average reply rate when sending to a mixed pool of mailbox providers; a Google-only list can run around 1 percent. (Reply rate benchmarks)
• Below 0.5 percent at 1,000 or more sends points to an infrastructure or list problem rather than a copy problem. (Reply rate benchmarks)
• 10 percent or more of replies being genuinely interested marks a good campaign, and 20 percent or more marks a great one. (Reply rate benchmarks)
• Below 5 percent interested-of-replies means the offer is the problem, not the copy. (Reply rate benchmarks)
• Under 500 sends per interested reply is really good, 500 to 1,000 is good, and over 1,000 means the variant is underperforming. (Reply rate benchmarks)
• 1 interested reply per roughly 580 sends is a strong enough signal to push volume toward that variant immediately. (Cold email A/B testing)
• 0.86 percent reply rate, 18 percent of replies interested, and around 17 percent of interested replies converting to booked meetings described one of our best-performing campaigns, despite its reply rate sitting below the generic industry target. (Reply rate benchmarks)

What are the key cold email deliverability statistics?

• 6 percent is the bounce rate threshold that matters, and only when sustained across two or more consecutive days; a single-day spike is noise that self-corrects. (Cold email deliverability 2026)
• Close to 100 percent of cold mail from new sending domains is rejected on policy by Mimecast-protected domains, regardless of address validity. (Deliverability 2026)
• 7 percent of one campaign's list was Mimecast-protected, yet those domains produced 77 percent of all its bounces. (Deliverability 2026)
• 35 percent was the bounce rate on Mimecast-protected domains in another campaign, a 16x over-index against the rest of the list, where they were around 4 percent of leads and 64 percent of bounces. (Deliverability 2026)
• 0.6 percent, 0.9 percent, and 1.2 percent were the bounce rates for Microsoft 365, Google, and Proofpoint in the same analysis, which is why mainstream providers are fine to send to and appliance-style gateways are not. (Deliverability 2026)

What are the key sending volume and sender statistics?

• 1 cold email a day is where every new sender starts, scaling to around 3 a day only once the campaign is producing interested replies. (Cold emails per day per sender)
• 2 to 3 weeks of warmup is required before any new sender account takes full cold volume. (Emails per day per sender)
• 2 sends per account per day is the fleet-wide average across a large mix of proving and proven senders. (Emails per day per sender)
• 3 sender accounts per domain is the standard structure, so no single account or domain carries enough weight to look industrial. (Emails per day per sender)
• 50,000 emails a month works out to roughly 2,500 sends a day, which needs on the order of 1,250 mailboxes at a 2-per-account average, spread across several hundred secondary domains. (Emails per day per sender)
• 1,000-plus is where campaign-level daily caps should sit, purely so they never become the binding constraint; we once found a campaign pinned at 999 sends a day while its senders could collectively handle nearly 1,600. (Emails per day per sender)

What are the key follow-up and sequence statistics?

• 4 steps is the right length for a cold sequence: the opener, a follow-up 3 days later, a completely new offer after a 30-day gap, and a final follow-up 3 days after that. (Cold email follow-up strategy)
• 3 days is the hard minimum gap between any two sequence steps, cold or warm, no exceptions. (Follow-up strategy)
• 8 further touches is the maximum we run after a prospect shows interest and goes quiet, landing at roughly days 3, 6, 10, 14, 18, 22, 26 and 30. (Follow-up strategy)
• 150 words is the ceiling on every follow-up email: one new angle, one case study result with a real number, one fresh call to action. (Follow-up strategy)

What are the key cold email testing statistics?

• Roughly 1,000 sends per variant is the floor before a variant read counts as settled rather than early. (Cold email A/B testing)
• Roughly 2,000 sends, about 20 replies, and zero interested replies is the evidence bar for declaring a variant or offer dead. (A/B testing)
• About 2x is the maximum send gap at which a comparative verdict between two variants is readable; beyond that, reject the call. (A/B testing)
• 7x was the interested-per-send density gap between the winning and losing variants in one four-variant test at comparable volume, around 0.141 percent versus 0.020 percent, and our review still declined to call it for a week because every variant had scored once. (A/B testing)
• 0.81 percent was the reply rate of the worst variant in that test, converting just 2.5 percent of replies to interested, while the winner replied at 0.63 percent and converted 22.6 percent, proving raw reply rate is a decoy in variant selection. (Dead campaign anatomy)
• Roughly 10,000 campaign sends is the floor for whole-campaign conclusions. (Dead campaign anatomy)
• Roughly 2,000 cumulative sends per variant is the point at which spintax must deepen, because past it the pattern goes stale and reply rates decay. (Spintax for cold email)

What are the key list quality statistics?

• 45 percent was the bounce rate on a campaign where we trusted a list's stale "valid" flags instead of re-verifying; only a small fraction of the addresses marked valid were still genuinely valid. (Email list verification guide)
• More than 50x is the price spread we found between email verification providers doing the same job at the same accuracy, so shop on price per check. (Verification guide)
• 38 percent of one list build was valid-risky catch-all addresses, a large enough slice of genuine buyers that dropping them all would have gutted the list; they get sent as a separate pool instead. (Verification guide)
• 100 percent of unique domains get MX-checked on every list, never a sample, because a clean spot check on one build missed protected domains that the full sweep then found. (Verification guide)

What are the key LinkedIn outreach statistics?

• Around 20 percent is how much messages under 150 characters outperformed longer ones in our data across 10,000-plus DMs. (LinkedIn outreach message patterns)
• Over 40 percent better is how sequences with 3-plus touchpoints performed against single messages. (LinkedIn message patterns)
• 55 percent higher reply rates come from signal-based messaging versus generic outreach. (LinkedIn message patterns)
• 18 percent-plus reply rates, with 8 percent-plus positive replies, is what well-tuned tier-one signal-driven LinkedIn campaigns hit. (LinkedIn message patterns)
• 30 percent or more is the jump in connection acceptance when warm touches like a profile view or post like precede the request, with warmed campaigns running 40 percent-plus acceptance. (LinkedIn message patterns)
• 14 percent is the additional lift in response rates from adding an email leg for LinkedIn non-responders, and the two channels coordinated outperform either alone by 30 to 40 percent. (LinkedIn message patterns)

Which cold email numbers should you memorise?

If you only keep a handful of these statistics, keep the ones that gate decisions. This table is the short list we would put on the wall.

Number	What it governs	Source
0.5-1.0%	Healthy raw reply rate at volume; the deliverability signal	Reply rate benchmarks
10% / 20%	Good / great share of replies that are genuinely interested; the offer signal	Reply rate benchmarks
~2,000 sends, ~20 replies, 0 interested	The kill bar for declaring an offer or variant dead	A/B testing
6% over 2+ days	The only bounce rate worth diagnosing; anything less is noise	Deliverability 2026
1 then ~3 per day	Per-sender cold volume, unproven then proven	Emails per day per sender
3 days	Minimum gap between any two sequence steps	Follow-up strategy
Under 150 characters	LinkedIn message length that outperforms by ~20%	LinkedIn message patterns

One closing warning. A benchmark only means something at the level it was measured: reply rate is a deliverability signal, variant numbers must be summed to offer level, and no number on thin volume is a verdict. The linked posts show the working; the numbers are the summary, not the substitute.

FAQ

What is the average cold email reply rate in 2026?

0.5 to 0.6 percent is average at volume across a mixed mailbox provider pool. 0.5 to 1.0 percent is the healthy baseline, 1.0 to 1.5 percent is good, and above 1.5 percent is really good. A Google-only list can run around 1 percent.

What is a good interested reply rate for cold email?

10 percent or more of replies being genuinely interested is good, and 20 percent or more is great. Below 5 percent means the offer is the problem, not the copy. Measured as density, under 500 sends per interested reply is really good and 500 to 1,000 is good.

How many cold emails can you send per day per inbox?

Start each new mailbox at 1 cold email a day after 2 to 3 weeks of warmup, and scale to around 3 a day only once the campaign is producing interested replies. Across a large fleet that averages out near 2 sends per account per day.

What is a bad bounce rate for cold email?

A bounce rate above roughly 6 percent sustained across two or more consecutive days is a real problem. A single-day spike is noise, because dead addresses bounce exactly once, get suppressed, and the rate self-corrects.

How many sends before a cold email offer is proven dead?

Roughly 2,000 cumulative sends to that offer, around 20 replies to it, and zero interested replies. Both floors must be met, and sends must be summed across every variant carrying the offer, because the offer is the unit under test.

Where do these cold email statistics come from?

From live client campaigns we run at Pipeline Playbooks, sending up to 50,000 cold emails a month per client. They are measured operating numbers, not survey responses. Auto-replies are excluded from stats, and every interested classification is verified by hand against the actual reply text.