How to pick send days and times
Find the send days and times that work for your own subscribers: start from your order data, test two times fairly, then use Klaviyo's send time tools once you qualify.
What this is for
There is no send time that suits every store. A shop selling lunch-break treats and one selling garden tools to retirees have different customers with different days. The honest way to pick a time is to start from your own data, test one time against another, and keep what wins for your list.
Klaviyo's own tools are built on that idea. Smart Send Time does not guess from a hidden formula; it sends at many times and learns from how your subscribers respond. source, checked 28 September 2026
Time needed and what you need
Time needed: about 1 hour to set up, then 4 to 6 weeks of testing. The setup is an order export and a test campaign; the answer comes from running tests on your normal campaigns over the following weeks.
- An export of the last 90 days of Shopify orders with the order date and time
- Your subscriber count, since some send time tools need a minimum audience
- Your campaign calendar for the next month or two
- A note of where your subscribers live, if you sell across time zones
The steps
1. Find when your customers already buy
Export recent orders from Shopify and count them by day of the week and hour of the day. A clear peak, such as Sunday evenings or weekday lunchtimes, is your first candidate send time. Send an hour or so before the peak so the email is near the top of the inbox when people sit down to shop.
2. Decide between local time and one fixed time
If your subscribers are spread across time zones, schedule campaigns in each recipient's local time zone, so 7 p.m. means 7 p.m. wherever they are. Profiles with no known time zone get your account's time zone instead. source, checked 28 September 2026
Use one fixed time for anything with a real deadline, such as a sale ending at midnight Eastern. Local-time sending spreads the campaign over the day, so the last time zone may get it after the deal has closed.
3. Avoid scheduling exactly on the hour
Many senders schedule at 9:00 or 10:00, so those minutes are busy. Klaviyo notes that sends at peak minutes can go out more slowly and suggests typing an odd minute instead. source, checked 28 September 2026
4. Test two send times on a normal campaign
Take a regular campaign, not a sale, and use the campaign A/B test to send the same email at two times. Keep everything else identical, or you will not know whether the time or the content made the difference. source, checked 28 September 2026
Use your order peak from step 1 as one time. If you have no idea for the other, Klaviyo's help centre suggests an evening slot as a starting point. source, checked 28 September 2026
5. Test days the same way, over several weeks
Days are harder to test inside one campaign, so rotate instead: send your weekly email on Tuesday for three weeks, then Sunday for three weeks, with similar content and the same time of day. Compare clicks and orders per recipient, not opens. Holidays and sales weeks distort the result, so leave them out of the comparison.
6. Run Smart Send Time once your campaigns reach 12,000 people
Klaviyo's Smart Send Time starts with exploratory sends that deliver one campaign across 24 hours in each recipient's local time, then narrows to a best hour. It only works for campaigns sent to a large audience. source, checked 28 September 2026
Keep to your usual sending rhythm while it learns. Sending extra campaigns to speed it up changes the behaviour it is trying to measure. source, checked 28 September 2026
7. Keep checking the winner with focused sends
Once Smart Send Time has an answer, its focused sends keep testing around it: one group at the chosen time and two groups two hours either side. If your customers' habits shift, say in summer, the chosen time moves with them. source, checked 28 September 2026
8. Consider Personalized Send Time if you have the add-on
Personalized Send Time picks a time for each subscriber inside a window you set, instead of one time for everyone. It needs one of Klaviyo's paid analytics packages and only works for campaigns scheduled for a later day, not flows. source, checked 28 September 2026
9. Write the result into your calendar
Once you have a winning day and time, make it the default in your monthly email calendar and space other campaigns at least a day or two apart. Re-run the time test each season; a free email calendar generator is a quick way to lay the month out around it.
A worked example
For a houseplant shop with 15,000 subscribers in the US and UK
Worked example (an invented shop, with invented numbers)
The shop's order export shows most orders between 7 p.m. and 10 p.m. on Sundays and Wednesdays. It tests its Wednesday newsletter at 10:04 a.m. against 6:04 p.m., in each recipient's local time, with a 50/50 split: 7,500 subscribers each (7,500 + 7,500 = 15,000).
The morning send gets 480 clicks and 61 orders. The evening send gets 585 clicks and 79 orders. That is 18 more orders from the same email, so the shop moves its Wednesday send to the evening.
Because its list is over 12,000, it then runs Smart Send Time. After four weekly exploratory sends the focused time lands at 7 p.m., and focused sends go out at 5 p.m., 7 p.m. and 9 p.m. local time to keep checking it.
Common mistakes
Using local-time sending or Smart Send Time for a flash sale. The campaign rolls out across the day, so a deadline arrives before some people have the email. source, checked 28 September 2026
Copying a "best time to send" from someone else's list. Another store's winning hour tells you about their customers. Run your own test; it costs one campaign.
Judging a time test on opens. Automatic opens from Apple Mail arrive whenever the email lands, whatever the hour, so they flatten the difference between times. Clicks and orders show whether people were ready to act.
Changing the send time and the subject line in the same week. When results change, you will not know which one did it. Settle the time first, then work on subject lines.
Doing this with Tilly
Tilly's email calendar agent plans a month of campaigns, and the plan itself costs no credits. Each campaign Tilly then writes costs 10 credits and goes to Klaviyo as a draft, so the day and time you settled on here is set by you when you schedule it in Klaviyo.
Questions people ask
My list is under 12,000. Can I still use Smart Send Time?
Not for the exploratory and focused tests. Use campaign A/B tests on send time instead, as in step 4, and repeat them a few times so one odd week does not decide it for you. source, checked 28 September 2026
How long after a send should I wait before judging it?
At least a couple of days, because orders keep arriving after the opens. Klaviyo's Smart Send Time model itself waits for the conversion period, five days by default, before it records a result. source, checked 28 September 2026
Should flows follow the same send time?
No. Flows fire from what a customer did, such as leaving a checkout, and their timing should follow that event. Send time tuning is for campaigns. For how often to email overall, see setting an email frequency cap.
See this on your own store
Paste your store URL. The first pass takes about a minute and needs no account. Save a card (nothing is charged) and Tilly reads the whole store and works out who buys from you.