TL;DR: Across 18,718 first replies to LinkedIn cold outreach, 51.5% arrived between 09:00 and 18:00 UTC, and the average weekend day carried 42% fewer replies than the average weekday.
Across 18,718 first replies to LinkedIn cold outreach, 51.5% arrived between 09:00 and 18:00 UTC, Tuesday carried more replies than any other day, and the average weekend day carried 42% fewer replies than the average weekday. Platform data
Key Takeaways
- 51.48% of all first replies arrived between 09:00 and 18:00 UTC.
- Tuesday carried more replies than any other day, 20.35% of the total.
- The average weekend day carried 41.98% fewer replies than the average weekday.
- The four hours from 13:00 to 16:59 UTC carried 28.44% of all replies.
- One acceptance-timestamp test failed and was discarded, which is why this study trusts reply timestamps instead.
What did we measure?
Most timing research in outreach asks when you should send. We measured that question and found almost nothing there. This study asks the mirror question, which turns out to have a real answer: when do the people you contacted actually reply?
Reachium records the timestamp of the first inbound reply on every outreach conversation. This study is the histogram of 18,718 of those timestamps, by hour of day and day of week. It describes recipient behavior, which no amount of send scheduling changes, and which anyone staffing an inbox is implicitly guessing at. Platform data
How was it measured?
Every timestamp in this study is a first reply recorded between 8 January 2025 and 27 August 2026, in UTC.
Dataset. Every recorded first reply to a connection-request campaign in the dataset, 18,718 timestamps, from requests sent between 8 January 2025 and 27 August 2026, pulled on 27 August 2026. Replies are conditional on having occurred, so this uses the all-sent basis rather than a maturity-restricted cohort. Hours and days are UTC. Platform data
Denominators. Each hour's share is that hour's replies divided by 18,718. Weekday and weekend daily averages divide each group's total (15,192 and 3,526) by its number of days (5 and 2). Platform data
A test this method had to pass first. Histograms of platform timestamps can measure the platform instead of the people. We ran the same histogram on acceptance timestamps and it failed exactly that way: 8,420 acceptances at 03:00 UTC against 21 at 17:00, an impossible human pattern produced by batch syncing, so acceptance timing was discarded entirely. Reply timestamps pass: the curve below is smooth, rises through the working day, and dips on weekends, which batch artifacts do not produce. One residual artifact survives and is flagged in the findings. Platform data
What did we find?
Reply volume by hour of day, in UTC, shows a clear midday concentration:
| Hour | Replies | Hour | Replies | Hour | Replies |
|---|---|---|---|---|---|
| 00:00 | 446 | 08:00 | 595 | 16:00 | 1,286 |
| 01:00 | 446 | 09:00 | 594 | 17:00 | 1,080 |
| 02:00 | 549 | 10:00 | 660 | 18:00 | 933 |
| 03:00 | 507 | 11:00 | 874 | 19:00 | 827 |
| 04:00 | 537 | 12:00 | 1,105 | 20:00 | 737 |
| 05:00 | 506 | 13:00 | 1,337 | 21:00 | 589 |
| 06:00 | 521 | 14:00 | 1,394 | 22:00 | 1,008 |
| 07:00 | 521 | 15:00 | 1,306 | 23:00 | 360 |
The peak hour is 14:00 UTC (1,394 replies) and the trough is 23:00 (360). The four hours from 13:00 to 16:59 carry 28.44% of all replies. The window from 09:00 to 17:59, nine of twenty-four hours, carries 51.48%. Platform data
One bucket breaks the evening decline: 22:00 UTC (1,008). That bucket is 37.20% one customer workspace, and no other hour shows that concentration. We read it as one large sender's audience in a distinctly later time zone, and flag it rather than smooth it. Platform data
By day of week: Monday 2,980, Tuesday 3,809, Wednesday 2,972, Thursday 2,881, Friday 2,550, Saturday 1,797, Sunday 1,729. Tuesday is the peak with 20.35% of all replies. The weekend share is 18.84%, and the average weekend day (1,763 replies) runs 41.98% below the average weekday (3,038). Platform data
What does this mean for inbox teams?
LinkedIn replies are an office-hours behavior. Half the volume lands in a nine-hour UTC window, Tuesday out-produces Sunday better than two to one, and the curve looks like a workday because that is what it is: people process their LinkedIn inbox as work, not as evening entertainment. Platform data
Note what this does not contradict: sending on weekends costs almost nothing even though replying on weekends is rare. A request sent on Saturday sits in the queue and gets answered on Tuesday. Send timing and reply timing are different clocks, and only the second one describes human attention.
The practical asymmetry is speed on the other side. The time-to-accept study reports that half of first replies arrive within 31 hours of acceptance. In this study's histogram those replies cluster inside working hours. A reply that lands at 14:00 and waits until the next morning for an answer has spent most of its freshness. Platform data
Teams that want inbox coverage handled for them rather than staffed in-house can compare options on Reachium's pricing page.
How do you use this in practice?
A founder running their own outreach on evenings and weekends is structurally mismatched with their repliers: 51.48% of replies land from 09:00 to 17:59 UTC, when that founder is doing their day job. The data does not say the meetings are lost, but the conversations are happening on a delay in exactly the window when most post-acceptance bookings are decided. Platform data
A team staffing inbox coverage can size it from the table: the 13:00 to 16:59 UTC block carries 28.44% of reply volume in four hours. For a mostly US and European audience, that is the overlap of the US morning and the European afternoon. Weekend coverage, by contrast, addresses 18.84% of volume spread over two days. Platform data
For a worked example on the same platform, read the Yanne Capital case study.
What are the limits of this study?
UTC-only timestamps and platform-side recording constrain what this study can claim.
UTC only. Recipient time zones are not recorded, so "09:00 to 18:00 UTC" is not anyone's local business hours. The concentration is real; its local-time interpretation is inference, anchored by a dataset that skews toward US and European targets.
Platform-side timestamps. A reply is stamped when the platform's inbox sync observes it. Sync lag shifts individual replies later by minutes to hours. The acceptance-clock failure described in Methodology is the worst case of this class; the reply curve's shape says it is not batch-dominated, but exact per-hour values carry sync noise.
First replies only. Ongoing conversation messages are not counted, so this describes when conversations open, not all inbox traffic.
Concentration. The largest workspace accounts for 22.88% of matured requests, and the 22:00 UTC bucket shows what that can do to a single cell. The aggregate curve should be read as the behavior of the audiences these teams target, not of LinkedIn's population. Platform data
No outcome link. This study counts when replies happen, not whether fast-answered replies convert better. Nothing here measures response-time effects. Teams wanting to measure their own response-time effect can talk to the agency team about setting that up.
FAQ
When do people reply to LinkedIn outreach messages?
51.48% of first replies arrive between 09:00 and 18:00 UTC, with a peak at 14:00 UTC, treating LinkedIn replies as an office-hours behavior.
Which day of the week gets the most LinkedIn replies?
Tuesday, with 20.35% of all replies, more than twice Sunday's share.
Do LinkedIn replies drop off on weekends?
Yes. The average weekend day carries 41.98% fewer replies than the average weekday, even though sending on weekends costs little in acceptance.
When should a team staff inbox coverage for LinkedIn replies?
The 13:00 to 16:59 UTC block alone carries 28.44% of reply volume. Teams that want that coverage handled for them can compare options on Reachium's pricing page.