Best Time to Send Cold Emails in B2B, Study by Study

The best time to send a B2B cold email is Tuesday to Thursday, early to mid morning, roughly 7 to 11 AM in the recipient's own time zone. That is where the credible studies overlap. Smartlead, a sending tool measuring its own customers' send patterns, puts the best hour at 7 AM recipient time, with a 1.34% reply rate against 0.82% at 4 PM, and the best day at Wednesday, at 0.91%.

What most timing articles skip: the day barely matters. In that same data, every weekday sits between 0.87% and 0.91%. The hour moves the number far more than the day, and the recipient's time zone decides what "the hour" even means.

I traced 9 timing studies that rank for this question. Only 4 describe both their sample and their date range. Several rank send times by open rate, a number Apple broke in 2021. Below is what each one measured, how far I would trust it, and what I would set as a default tomorrow. All sources checked September 2026.

What day of the week gets the most cold email replies?

Wednesday wins most often. The margin is tiny.

Five of six put Wednesday at or near the top. Now look at the size of the win.

In Smartlead's numbers, Wednesday beats Friday by 0.04 percentage points. On 10,000 emails that is 91 replies against 87. In the agency study, Wednesday and Thursday sit at 0.48% while Monday and Tuesday sit at 0.45%. Three replies per 10,000 emails.

A founder sending 2,000 emails a month would never see that difference through normal noise. So "which day?" is the least useful timing question. Instantly's 2026 benchmark report, built on its own users' sends, adds one practical detail: Friday brings the most auto-replies, which fits the slightly weaker Friday numbers elsewhere.

Saturday is the one day that loses in both studies that report weekend reply rates. Smartlead puts it at 0.82%, its lowest day, and the agency study at 0.35%, also its lowest. Sunday is where they split. Smartlead has Sunday at 0.90%, level with Tuesday, and its advice is to send "on any weekday or Sunday." The agency study shows Sunday with its highest raw rate, 0.57%, but on very low volume, driven by a small group of senders who time their emails for executives catching up on Sunday evening.

I would not build a default on either Sunday number. Smartlead does not disclose its sample, and the agency's Sunday rests on a sliver of its sends. Sunday is worth a small test. Saturday is the day I would skip.

What time of day should a cold email land?

Morning, in the recipient's clock. Every reply-based source that names an hour picks a morning slot.

The hour gap is real and larger than the day gap. In Smartlead's data, 7 AM against 4 PM is 134 replies against 82 per 10,000 contacts. That is 52 more replies from a scheduling setting.

Two details change how to read this table.

The 2025 agency study logged the moment each email was sent. Its own methodology note warns that for audiences across several time zones, the recipient's local time may differ. So its "8 AM to 12 PM" is partly the sender's morning.

The 2021 study counts share of replies, a different measure from reply rate. "20% of replies came from 8 AM sends" only means 8 AM won if the same number of emails went out every hour, and the page does not publish send volume by hour.

A team that sends most of its emails at 8 AM will collect most of its replies at 8 AM. The 2025 agency study shows why this matters: afternoon carried by far the most send volume in its data, and still trailed both morning windows on rate.

Why do cold email timing studies disagree?

Because they are rarely measuring the same thing. Across the 9 sources, I found four different yardsticks:

• Open rate: the share of emails a tracking pixel says were opened
• Reply rate: replies divided by emails sent, or by people contacted
• Share of replies: the percentage of all replies that came from one send slot
• Meeting rate: booked meetings divided by emails sent

They also use different clocks (sender or recipient), different populations (one agency's clients or every sender on a platform) and different years, from early 2021 to 2025. Put those side by side without saying so, and you get pages where Tuesday, Monday or Thursday wins, depending on whose blog you open.

The open rate problem nobody names

Apple's Mail Privacy Protection shipped on September 20, 2021, as part of iOS 15. Apple's own announcement says it "prevents senders from learning whether an email has been opened."

Apple's policy page explains the mechanism: Mail "downloads remote content in the background by default," whether or not the reader ever engages with the email, and routes it through two relays so the sender never sees the reader's IP address.

The tracking pixel your sending tool counts as an "open" is remote content. When Mail fetches it in the background, the open gets recorded whether or not a person read anything. The timestamp then records when a machine fetched an image. When, or whether, a buyer read your email stays unknown.

For a timing study, that is fatal. An "opens by hour" chart becomes partly a chart of background downloads.

Now go back to the hour table. The 10 AM claim with a 38% open rate and the "30-40% higher open rates" morning claim both rest on opens. None of the 9 sources I read names Mail Privacy Protection. Only the 2025 agency study addresses the problem at all. It stopped dividing by opens and started dividing by sends, saying email clients and privacy tools now block too many tracking pixels to trust the count.

My rule for reading any timing study: if the claim is built on opens, skip it. Replies are better. Booked meetings are better still.

Which timing studies can you trust?

Here is the grading I used. Four questions per source: is the sample stated, is the period stated, what was measured, and what is the main weakness.

Every source here comes from a company that sells outbound services or sending software. That includes the category I work in, and I have no timing study of my own to add, so this page grades other people's work. Weight each row by what it discloses, and the picture gets simple. The sources that disclose the most agree on morning in the recipient's time zone, midweek by a hair, and Saturday as the weakest day wherever weekends are reported.

Does the best time for replies also book the most meetings?

Not always, and this is the finding I would build a schedule around. Only 1 of the 9 sources reports booked meetings: the 2025 agency study. Its meeting rate is booked meetings divided by emails sent.

Two things jump out.

First, the reply gap between morning and afternoon is small (0.54% against 0.47%), but the meeting gap is 4x (0.4% against 0.1%). Say you send 4,000 emails a month. On these averages, morning gives you about 22 replies and 16 meetings. Afternoon gives you about 19 replies and 4 meetings. Three fewer replies, twelve fewer meetings.

Second, Wednesday ties Thursday for the highest weekday reply rate and has the lowest meeting rate of the week. The study reads that as more "not interested" and more one-line brush-offs. If you rank send times by replies alone, you can end up tuning your schedule for polite rejections.

Read the 4x with care. Both meeting rates are rounded to one decimal. A 0.1% could be anywhere from 0.05% to just under 0.15%, and a 0.4% anywhere from 0.35% to just under 0.45%, so the true gap sits somewhere between about 2x and 9x.

Taken literally, the morning figures would also mean most morning replies became meetings. The study does not say how a meeting booked after a follow-up gets assigned to a send window, which may explain it. Treat the finding as a strong direction: morning sends book more meetings. How many more on your own calendar, this study cannot tell you.

Smirnov Consulting Group is a Prague-based B2B outbound lead generation agency that runs cold email and LinkedIn campaigns for founder-led B2B companies and books qualified sales calls. Booked calls are the number my work gets judged on, so the column I read first in any timing study is the meeting column. In 8 of these 9 sources, it does not exist.

How do you send in the recipient's time zone?

Almost every source that measured hours used the recipient's clock, and the one that did not warns about it. Sending on your own clock is the easiest timing mistake to make, because it happens by default.

What it looks like from Prague. A batch scheduled for 9 AM here reaches London at 8 AM, which is fine. It reaches New York at 3 AM on most days of the year, where it sits for hours under everything else that arrives before the prospect's day starts. Send the same batch to a US list at 9 AM Eastern, and a buyer in California gets it at 6 AM.

The fix is mechanical:

1. Put a time zone on every lead before launch. Use the person's location if you have it, the company's head office if you do not.
2. Split the list by zone, or use per-lead scheduling if your sending tool supports it.
3. Set the window to 7 to 11 AM local, Monday to Friday.
4. For US lists, check the zones separately. 7 AM Eastern is 4 AM Pacific.

There is a simpler version. Target a local audience when you can. It is easier to open a conversation with people who know you live in the same city. You also share their working day, so the send-time question mostly answers itself.

Why won't the perfect send time fix a weak campaign?

Look at the size of the effects in this article. The biggest day gap in any reply-rate study here is a fraction of a percentage point. The hour gap Smartlead reports, 7 AM against 4 PM, is 0.52 points.

Now compare that with what changes when the sender changes. One cold email, word for word the same, earned a 2% reply rate from one B2B company's outbox and 7.5% from another's. That is a gap of 5.5 points, about ten times the largest timing gap on this page.

Different campaigns and different denominators, so read it as an order of magnitude. The order of magnitude is the point.

Send time is a multiplier on a good list and a credible sender, not a fix for a weak one. A reply rate, as that post argues, "belongs to whoever sent it."

The list matters the same way. I have written it before: "Lists bought cheap from freelancers produce low open rates and high bounces." The bounces alone tell you the problem, and no send window changes who is on the list. If your numbers are poor, audit the list first and the schedule second.

Get the list and the sender right first. Then timing earns you the last few points, and those points are worth having.

What should you set as your default send time?

Here is what I would set tomorrow for a B2B campaign, based on the sources that disclose the most:

1. Days: Monday to Friday. Midweek wins by a hair, too small to justify idling your inboxes on Monday and Friday. Skip Saturday. Sunday scores well in Smartlead's data, but on an undisclosed sample, so add it only as a deliberate test, such as a small Sunday evening send to executives.
2. Hours: 7 to 11 AM in each recipient's time zone, weighted toward the start of that window.
3. Scoring: judge every window on replies and booked calls. Ignore opens for timing decisions.
4. Testing: test morning against afternoon and leave Tuesday against Wednesday alone. The day gap is too small to detect, and even an hour test needs several thousand sends per version before the result means anything.
5. Follow-ups: keep the same local morning window for every step. Instantly's timing guide recommends 3 to 4 days between touches, and its benchmark data shows 58% of replies come from the first email, so the first send deserves your best slot. For how many steps to run and what goes in each, see my breakdown of a follow-up sequence.

One more reason to trust a morning test quickly. In the 2021 SDR firm data, 87% of replies arrived within 24 hours of the send. You will know within a day whether a window is pulling, even if you need weeks of volume to be sure.