Send between 8am and 11am in the recipient’s local timezone, Tuesday through Thursday. Getting the timezone right matters more than getting the hour right, and both matter less than whether the message was relevant.
Published by emailcampaign.ai. This page argues that a popular topic is mostly a distraction, which is not the usual incentive for writing about it.
The consensus, and its size
| Day | Assessment | Reasoning |
|---|---|---|
| Monday | Below average | Competing with the weekend backlog |
| Tuesday | Good | Backlog cleared, week not yet full |
| Wednesday | Good | Consistently the steadiest performer |
| Thursday | Good | Still inside the working week |
| Friday | Below average | Attention already on next week |
| Weekend | Poor for B2B | Buried under Monday's arrivals |
Within a good day, mid morning works because people triage their inbox after settling in. Early afternoon is a weaker second window. Late afternoon performs poorly because messages arriving then get deferred and then forgotten.
But hold the size of this effect in mind. Moving a campaign from Friday afternoon to Wednesday morning might shift reply rate by a fraction of a percentage point. Narrowing the target list can double it.
Why the studies overstate it
Nearly every send time study measures open rate, and open rate stopped being a clean signal in 2021 when Apple Mail Privacy Protection began pre-fetching images on behalf of recipients.
The distortion is specifically bad for timing research. A proxy fetch happens close to delivery, not close to reading. So a message delivered at 9am and actually read at 4pm is recorded as a 9am open. That does not add noise evenly; it manufactures a correlation between send time and apparent open time, which is exactly the relationship the study claims to have found.
If you want to test timing for yourself, test against reply rate, which requires a person.
Timezone handling is the part that matters
A perfectly chosen 9am send is worthless if the recipient is eight hours away and it lands at 1am. Their morning triage happens long after your message has been buried.
If your tool supports timezone aware scheduling, use it. If it does not, segment the list by region and schedule each segment separately. For most B2B lists three buckets are enough: Americas, Europe and Middle East, and Asia Pacific.
Where you cannot determine a timezone, the company’s headquarters location is a better guess than your own.
The timing question that does affect deliverability
Not the hour, the burst. A mailbox that sends its entire daily allocation in sixty seconds looks like software, because it is. A mailbox that sends eight messages spread across a working day with uneven gaps looks like a person.
So the settings worth checking are the ones about distribution rather than start time: spread sends across several hours, randomise the interval between them, and keep the daily total per mailbox at eight to ten. That pattern is a genuine deliverability input, unlike the choice between 9am and 10am.
Where to spend the attention instead
Timing is the last variable to optimise, after the ones that decide whether the message arrives at all: authentication that resolves rather than merely exists, a hard bounce rate under two percent, a complaint rate under Google’s published 0.1 percent, and sensible volume per mailbox.
Set sending to mid morning Tuesday to Thursday in the recipient’s timezone, spread across the day, and then stop thinking about it.
Questions, answered straight
- When is the best time to send cold emails?
- Mid morning in the recipient's local timezone, roughly between 8am and 11am, on Tuesday, Wednesday or Thursday. The effect is real but modest, and it is smaller than the difference made by targeting or deliverability.
- What is the worst day to send cold email?
- Friday afternoon and the weekend. A message arriving Friday at 4pm is competing with someone finishing their week, and a message arriving Saturday is buried under Monday's arrivals before it is seen.
- Does send time affect deliverability?
- Not directly, but sending your whole day's volume in one burst does. Spreading sends across the working day with random intervals looks like a person writing emails. Two hundred messages leaving one mailbox in a minute does not.
- Should I send in the recipient's timezone?
- Yes, and it matters more than the exact hour. A carefully chosen 9am send is worthless if it lands at 2am for the recipient. If your tool does not support timezone aware sending, segment the list by region and schedule separately.
- Do send time studies actually prove anything?
- Less than they claim. Most measure open rate, which has been partly machine generated since Apple Mail Privacy Protection launched in 2021, and automated fetches happen at delivery time rather than reading time. That systematically distorts any timing conclusion.
Published 20 September 2026. Updated 20 September 2026. Written by the team that runs the infrastructure; numbers come from the platform's own provisioning and sending, and from the providers' published documentation at the time of writing.