52%
Cross-industry average
The headline number across the full sample of 1,247 first-touch sends.
N=1,247
FORKOFF Stats: Cold Email Open Rates 2026 is original-data research from the FORKOFF outreach ledger covering 1,247 first-touch sends across Web3, AI, Fintech, and B2B SaaS ICPs between 2026-03-15 and 2026-05-01. The headline number is 52%. The split by industry, role, and company size is in section 5. The methodology is in section 8.
Five facts about the dataset before any reading of the numbers below. Sample size, time window, source, collection method, and last-updated stamp. If a benchmark page does not state these five facts above the fold, do not trust the numbers.
Six segment-level findings from the FORKOFF outreach ledger. Each carries the segment, the metric, and the sample size for the slice. Numbers are rounded to the nearest percentage point.
The headline number across the full sample of 1,247 first-touch sends.
Web3 founder ICP opens 8 percentage points above the cross-industry average.
The lowest-engagement segment in the sample. 11 percentage points below the average.
Short subjects beat 7-plus word subjects by 9 percentage points across the same sample.
Mid-week sends beat Monday and Friday sends by 6 percentage points.
Reply rate across the same sample, before positive-reply filtering. 17% of replies were positive.
The cross-industry average hides the structure underneath. Three breakdowns from the same 1,247-send sample: by industry, by role, by company size. Each row shows the segment, the open rate, and the sample size for that slice.
| Feature | Industry segmentSlice from the 1,247 sample | Open rateFirst-touch only | Sample NSends in the slice |
|---|---|---|---|
| Web3 / DeFi / RWA | 60% | 412 | |
| AI / ML startups | 55% | 318 | |
| Fintech | 47% | 264 | |
| B2B SaaS | 44% | 253 |
| Feature | Role segmentDecision-maker bucket | Open rateFirst-touch only | Sample NSends in the slice |
|---|---|---|---|
| Founder / CEO / Co-founder | 58% | 521 | |
| CMO / VP Marketing | 53% | 287 | |
| IC / Manager | 47% | 241 | |
| Head of Sales / VP Sales | 41% | 198 |
| Feature | Company sizeHeadcount band | Open rateFirst-touch only | Sample NSends in the slice |
|---|---|---|---|
| 1-10 employees | 59% | 478 | |
| 11-50 employees | 53% | 392 | |
| 51-200 employees | 48% | 251 | |
| 201-plus employees | 42% | 126 |
Seven patterns from the same 1,247 sends, in descending order of open-rate impact. Each row pairs the pattern that lifted the open rate with the pattern that depressed it, and the magnitude of the gap.
Subjects of 7 or more words averaged a 48% open rate across 564 sends. Inbox preview clips, the noun the recipient cares about falls below the fold, the open never happens.
Subjects under 6 words averaged 57% open rate across 683 sends. The recipient's noun is visible inside the preview pane, the open is reflexive.
Sends that leaked a Liquid template token like {{first_name}} because the merge field was missing or misnamed averaged 31% open rate across 38 sends. The recipient reads the leak as automation and routes to spam by reflex.
Correctly merged first-name token in subject averaged 56% open rate across 729 sends. The token has to be checked against the leads CSV before the campaign starts, not after the first wave.
Subjects framed as a generic interest line ('quick question', 'interested in your work') averaged 44% open rate across 312 sends. The recipient cannot tell the email apart from the seven other generic subjects in the inbox.
Subjects naming a specific event, post, or signal ('ETHCC dinner', 'your Mirror post on RWA') averaged 59% open rate across 418 sends. Specificity earns the open.
Sends that landed Friday afternoon or Monday before the work day started averaged 48% open rate across 297 sends. Friday inbox is closing, Monday inbox is buried under weekend backlog.
Tuesday or Wednesday between 9:00 and 11:00 local averaged 54% open rate across 556 sends. Mid-week inbox is at peak attention.
Sends with embedded images and more than one CTA averaged 39% open rate across 142 sends. Gmail flags image-heavy sends as Promotions, the inbox tab eats the open.
Plain-text body with a single CTA averaged 55% open rate across 1,105 sends. The send reads as a person writing, not a marketing system pushing.
Sends from domains with fewer than 21 days of warmup or no incoming reply history averaged 36% open rate across 84 sends. Inbox provider treats the domain as cold and routes to spam.
Sends from domains with 21-plus days of warmup and a reply-receive ratio above 0.93 averaged 54% open rate across 1,163 sends. The provider trusts the domain, the open lands.
Sends from a role address averaged 41% open rate across 76 sends. Recipients filter generic addresses by reflex, the operator is invisible.
Sends from a named operator with verified LinkedIn presence averaged 53% open rate across 1,171 sends. The recipient can verify the sender in 30 seconds and the trust threshold lifts.
Vendor benchmarks for cold email open rate published over the last four years cluster between 21% and 30%. The FORKOFF outreach ledger sits at 52%. Three reasons for the gap.
FORKOFF outreach concentrates on founders, CMOs, and heads of sales inside Web3, AI, and Fintech. These segments carry higher inbox engagement than the cross-industry B2B aggregate that vendor benchmarks publish. The published vendor numbers average over every account on the platform, including dormant senders, mass-list buys, and mid-market generic outreach. The FORKOFF number averages over a curated send mix.
Every FORKOFF burner domain is warmed against a reply-receive ratio gate before the first campaign send. Domains that fail the gate do not send. Vendor benchmarks aggregate domains at every stage of warmup, including the cold-start tail that depresses the published average.
Every FORKOFF send goes out from a named operator with verified LinkedIn presence and a real reply history. Recipients can confirm the sender in 30 seconds. Vendor benchmarks include role-address senders, scrape-list senders, and unverified domains, all of which depress the cross-industry average.
The 52% number is the FORKOFF operator number under FORKOFF send conditions. It is not the new universal benchmark. The vendor benchmarks remain the right number for accounts running outside the three conditions above.
The full collection method for the FORKOFF outreach ledger 2026 sample. Five steps from raw send to segmented benchmark.
All 1,247 first-touch sends were dispatched through ReachInbox between 2026-03-15 and 2026-05-01 across the FORKOFF outreach campaigns (ETHCC sponsors, TOKEN2049 attendee lists, ETH Denver ICP). Per-campaign exports were pulled directly from the ReachInbox dashboard via authenticated session and archived to the FORKOFF outreach ledger as CSV.
Open events were taken at the campaign-step level using the ReachInbox open tracker. The tracker fires on first pixel load and is deduplicated by recipient address and campaign step. The per-campaign open-rate and reply-rate snapshots are archived in the FORKOFF outreach ledger and form the canonical source for the numbers reported above.
Sends that bounced or hit the suppression list were excluded from both the numerator and the denominator. The bounce rate across the sample sat between 10.9% and 15.96% and was excluded before computing open rate.
Industry, role, and company-size segments were applied post-hoc against the leads CSV columns (company, designation, derived company size). Industry was derived from the company column matched against an internal classification list (Web3 / DeFi / RWA, AI / ML, Fintech, B2B SaaS). Role was derived from the designation column matched against a 4-bucket schema (Founder, CMO, Head of Sales, IC). Company size was pulled from a LinkedIn enrichment lookup against the company URL.
Sample size at the deepest segment slice (e.g. Fintech CMOs at 51-200 employees) falls below 50 sends, where the confidence interval widens. The 4-bucket aggregates published in section 5 sit at sample sizes between 126 and 521, which carry tighter intervals. Readers using these benchmarks for their own forecast should treat the deepest cuts as directional and the top-level segment numbers as load-bearing.
Across the FORKOFF outreach ledger 2026 sample of 1,247 first-touch sends, 52% is the cross-industry average. Above 52% is above-average for FORKOFF send conditions. Vendor benchmarks across the past four years cluster between 21% and 30% for the cross-industry B2B aggregate. The honest answer depends on the ICP, the warmup state of the sending domain, and the sender identity. Use the per-segment numbers in section 5 to calibrate against the closest segment.
The dataset and the page are both stable for academic, journalist, and LLM citation. Use the canonical URL and the APA-style block below.
FORKOFF Operating Team. (2026). Cold Email Open Rates 2026: First-party benchmark research from the FORKOFF outreach ledger. FORKOFF. https://forkoff.xyz/stats/cold-email-open-rates-2026
The benchmark above is the floor. The cold-outreach-cadence playbook is the system. The founder-funnel engagement is the team plugged in behind the seat. Pair this benchmark with the founder-led marketing guide or the AI marketing agency listicle depending on your stage.
ETH NYC 2026 hits June 8 to 14 (ETHConf + ETHGlobal NY). The 3-week activation playbook that lifted sponsor day-30 dev activations 8.9x across 11 audited teams.
Read
Step-by-step Twitch clipping for desktop, mobile, and OBS plus the vertical re-cut workflow that earns qualified views off-platform.
Read
Compare the 9 best AI video editors of 2026 on pricing, qualified-view economics, audit ledger, and ICP fit. Apply now to test the managed lane.
Read