The Best Time to Send Cold Emails: What Timing Changes, What It Does Not, and How to Measure Yours

Why the Charts Have No Source

Every article on this topic, including the one this page replaces, carried a chart: Tuesday 95%, Wednesday 92%, Friday 65%. Of what, measured where, by whom, on which audience, it never said, because the numbers were copied from the last article, which copied them from the one before. Timing effects exist, but they are small, they depend on who you are writing to, and they do not survive being turned into a universal percentage. We send more than a million cold emails a month and we still would not publish a weekday table, because our audience is cold email operators and yours is not.

What Timing Actually Changes

• •Attention. A message that arrives while the reader is clearing their inbox gets read in context; one that arrives at 11pm is forty messages deep by morning. Working hours in the recipient's zone is the one timing rule worth treating as fixed.
• •Pattern. Fifteen messages from one inbox at 09:00:00 is what a script does. Fifteen messages between 8:10 and 4:50 with uneven gaps is what a person does. Receivers see the pattern, and this is the timing decision that touches deliverability rather than replies.
• •Reply speed. Replies to morning sends come the same day; replies to late-afternoon sends come tomorrow. It changes when your pipeline moves, not whether.

What Timing Does Not Change

Whether the message lands. A perfectly timed email in spam has a reply rate of zero, and most campaigns that come to us asking about send times have an infrastructure problem they have not measured yet. Whether the list is right: a message to the wrong person at the right time is still the wrong message. Whether the copy is readable: length, a verifiable observation and one clear question move replies far more than the hour does. Fix those first, in that order, and treat timing as the last few percent.

Defaults That Are Safe to Start With

• •Sending window: 8am to 5pm in the recipient's time zone. Segment the list by zone if it spans more than two.
• •Gaps: randomised, 20 to 80 minutes between messages from the same inbox.
• •Days: Tuesday to Thursday for first touches; follow-ups can run Monday to Friday.
• •Weekends: off, for business addresses.
• •Follow-up spacing: two to four days, same thread, so the follow-up arrives on a different weekday from the first touch.

The Two-Week Test That Gives You Your Own Answer

Pick one campaign with one template and one list segment, so timing is the only thing that varies. Split it into send slots across the working day and across Tuesday to Thursday, and keep everything else identical. After two weeks, tag every reply as positive, neutral or negative and compute positive reply rate per slot; raw replies include out-of-office and unsubscribes and will mislead you. Keep the slot that wins by a margin you would bet on, and re-run the test when the list or the offer changes, because the answer changes with them. The positive reply rate article explains why that is the metric, and the tracking article shows how to cut the same data by domain and recipient provider, which usually turns out to matter more than the hour.

When Timing Looks Like the Problem and Is Not

If replies dropped and the send schedule did not change, the schedule is not the cause. Check the things that do change on their own: bounce rate creeping up as the list ages, a tracking domain shared with other senders, a domain's reputation wearing out, a pace that someone raised for a push. The spam diagnosis is the order to check them in, and every step has a free tool. Timing is the lever to pull once those are clean, and it is a lever worth a few percent, which on a fleet that sends a thousand a day is still worth an afternoon.