When is the best time to send cold email in 2026? Data from 750K sends
Most "best time to send" advice is recycled B2C marketing wisdom. Here is what cold-email reply data actually shows for B2B outbound.
Rejwan Nirob·May 23, 2026·6 min readThe internet is full of best-send-time advice that is half-true. "Tuesday at 10am" is the most-repeated answer and it is more right than wrong - but the why matters more than the headline. Pulled from our first 6 months in production: roughly 400 mailboxes, ~750K cold sends, reply-rate data measured at the 10 to 20 per day cap.
The two windows that actually matter
Window 1: Tuesday through Thursday, 9:00 to 11:00 in the recipient's local timezone. This is the highest-volume reply window across the entire dataset. Inboxes are checked first thing in the morning, recent enough to feel current. Reply rate in this window is roughly 1.4x the all-week average.
Window 2: Tuesday through Thursday, 13:30 to 15:00 in the recipient's local timezone. Smaller but still positive. Post-lunch inbox-checking. Reply rate runs about 1.1x the average. Worth adding to your sending schedule if Window 1 is fully booked.
The windows to avoid
- Monday all day - inboxes are catching up on weekend mail and prioritising existing threads
- Friday after 14:00 local time - the dead zone, weekend mode has kicked in
- Anything outside 8:00 to 18:00 local time - mail sent at odd hours pattern-matches to automation
- Weekends - sent volume is low but so is open rate, and any opens that do happen are skim-only
Why local timezone matters more than your timezone
Most cold-email tools send on the sender's timezone by default. This is wrong if your recipients span multiple regions. Sending at 10am London time hits 5am New York and 2am San Francisco - both wasted. Tools like Smartlead, Instantly, Lemlist, and Apollo all support per-recipient local-time scheduling. Use it.
The follow-up timing question
Follow-ups should arrive 2 to 4 working days after the original. Sooner is too pushy; later loses the context. The reply-rate curve we see across follow-ups: send 1 captures 40% of total replies, send 2 (day 3-4) captures 25%, send 3 (day 7-9) captures 20%, send 4 (day 12-14) captures 10%, send 5 onward captures the remaining 5%. Beyond send 5, marginal reply rate drops sharply and unsubscribe rate climbs.
“Tuesday morning local time wins the reply-rate chart. Friday afternoon and Monday all-day lose it. Everything else is variance.”
What does NOT matter as much as people think
Subject-line A/B tests inside the same send-time window are noisy. The signal that actually moves reply rate is the same five levers always: infrastructure quality (clean DNS), list quality (verified addresses), targeting (right ICP), opener line (specific to the prospect), and the call-to-action (one verb, one ask). Send time is a small multiplier on top of those.
Inboxlee provisions clean infrastructure so reply-rate variance comes from your copy and your list, not your sending stack. The send-time decision is yours; the placement that makes the send-time matter is ours.
See infrastructure monitoringFrequently asked
What is the best time to send cold email in 2026?
Tuesday, Wednesday, or Thursday between 9:00 and 11:00 in the recipient's local timezone. Reply rate in this window runs about 1.4x the all-week average. Secondary window: same days, 13:30 to 15:00 local. Avoid Monday all day, Friday after 14:00, and anything outside 8:00 to 18:00 local time.
Does the day of the week really matter for cold-email reply rates?
Yes, measurably. Tuesday through Thursday outperforms Monday and Friday by a meaningful margin in cold-email reply data. The drivers are inbox attention - Monday is catch-up mode for weekend mail, Friday afternoon is checked-out mode. Mid-week mornings catch recipients in their highest-attention state.
Should I schedule cold email in my timezone or the recipient's timezone?
Recipient's local timezone, always. Sending at 10am from London hits 5am New York and 2am San Francisco - both wasted. Smartlead, Instantly, Lemlist, and Apollo all support per-recipient local-time scheduling. Use the feature; the reply-rate uplift is meaningful for any multi-region recipient list.
How long should I wait between cold-email follow-ups?
Two to four working days between sends. Reply-rate distribution across follow-ups: send 1 captures 40%, send 2 (day 3-4) captures 25%, send 3 (day 7-9) captures 20%, send 4 (day 12-14) captures 10%. Beyond send 5, marginal reply rate drops sharply and unsubscribe rate climbs.
Does send time matter more than copy or list quality for cold email?
No. Send time is a small multiplier on top of infrastructure quality, list quality, targeting, opener line, and CTA. Sending the right message to the right person at 2am still beats sending the wrong message at the optimal time. Send-time optimisation is the last lever to pull, not the first.