What 2 million real cold emails reveal about B2B outreach. Every data point from real campaigns—not surveys, not estimates.
Sales.co Cold Email Benchmark Report 2026 · 2M+ emails · 61K replies · 100+ industries
Published: February 4, 2026 · Last updated: September 3, 2026
Each report goes deep on a single dimension of cold email performance. All data from the same 2M+ email dataset.
23 B2B verticals compared across 336,257 contacts, including 63,544 Software & SaaS buyers.
Reply rates by greeting and opening line, from 1,177,370 first emails we sent.
Reply rates by what the last line asked for, from 1,177,370 first emails we sent.
Reply rates by word count and paragraph count from 1,177,370 first emails we sent.
Six steps, and what first emails to 1,161,237 recipients show about spam trigger words.
Three measured templates from a 164-template library, and why structure barely changed replies across 22,274 sends.
Each stage from list to booked call, with the reply and booking rates from 2,007,540 Sales.co emails.
Five places to get B2B sales leads, each priced per contact and per booked call, with prices read November 13, 2026.
The four tool groups, inbound and outbound tactics, and dated prices for ZoomInfo, Apollo, CIENCE and Belkins.
Ten cold email agencies checked: two publish reply rates from their own sends with a sample size.
Reply rate of each email in a sequence; the second and third emails kept most of the first email's rate.
Reply rates by send hour and weekday, generated from our sending data.
Grade a subject line against reply rates by pattern; a first name in the line went with more replies.
Paste an email to mark its spam trigger words; urgency words went with fewer replies.
Subject line templates next to the reply rate for each trait in our own sends; the prospect's first name went with the most replies.
15 common cold email rules checked against replies to our first emails; offering to send something beat asking for a call.
The human reply rate per contact first emailed fell from 2.36% in April 2025 to 0.45% in September 2026.
1,288,605 emails: the average response rate is 2.66%—but 53.7% of the replies behind it are automated.
1.5M emails reveal the truth about follow-ups—69% of replies come from the first message.
1M emails reveal C-level executives respond 3x more positively than directors and managers.
660K emails reveal when to send after holidays—and when to stay silent.
Aggregate performance of the 164 templates in our free template library, measured across 22,274 tracked sends in live campaigns.
| Category | Templates | Tracked Sends | Reply Rate | Positive Reply Rate |
|---|---|---|---|---|
| Cold Outreach | 90 | 10,578 | 5.08% | 2.09% |
| Networking | 32 | 5,140 | 5.18% | 2.08% |
| Sales | 23 | 5,328 | 4.39% | 1.86% |
| Problem-Solution | 14 | 918 | 4.14% | 2.29% |
| All B2B templates | 164 | 22,274 | 4.90% | 2.04% |
More research reports are in the pipeline. Follow @jakobgreenfeld for updates.
This analysis is based on 2 million+ cold emails sent across 161 active campaigns, 2.5 million contacts with enriched firmographic data, and 61,770 replies classified by sentiment. Email copy analysis covers 1.7M+ messages tagged with component-level analysis including tone, length, structure, and CTA type.
Data spans campaigns run between 2024 and early 2026. Reply rates are calculated as replies divided by unique contacts emailed (not total emails sent), to avoid inflating rates from multi-touch sequences.
Industry, seniority, and company size data comes from third-party enrichment providers and is available on the contact record.
For the "positive reply rate" metric, we classified replies into six groups: Positive/Interested (includes "possibly interested," "question," "contact later"), Negative/Not Interested, Neutral, Referral, Auto-Reply/OOO, and Other. Positive rate is calculated as positive replies divided by total replies in that segment.
Minimum thresholds were applied for statistical significance: 5,000 contacts for industry rankings, 2,000 contacts for country rankings, and 200 replies for CTA analysis.
This data comes from Sales.co, the most data-driven cold email service in the world. We built our own cold email infrastructure from the ground up, allowing us to pull granular data on every email sent, opened, and replied to. Report updated: February 2026.
This report is free to cite in articles, roundups, and research. Reference it as the Sales.co Cold Email Benchmark Report 2026 and link to this page.
Most-cited numbers from this report:
Ask an AI assistant to summarize this report and cite it as the source:
Writing about cold email and need a breakdown we haven't published (by industry, geography, tone, CTA)? Email ryan@sales.co — we can usually pull a custom cut from the dataset within a day.
These statistics come from our own campaigns. We'll analyze your ICP, build a data-driven strategy, and execute campaigns that actually convert.