The best day to send a press release: we went and read the studies

The standard advice is that you should send a press release on a Tuesday, Wednesday or Thursday, in the morning, Eastern time. It is repeated on every distributor's blog, usually with a citation.

We went and read the cited studies. None of them measures what the advice claims, and the most rigorous of them says so itself.

What the studies actually measured

Prowly publishes the most-cited recent figure. Its methodology, in full, is one sentence: "We analyzed over 55470 press releases sent with Prowly's outreach tool to find out." The metric it reports is email open rate. The page does not state the time period, the sample composition, any control variables, or a margin of error.

Two problems. An open rate on a pitch email sent through one vendor's tool is not a measure of press-release performance — it measures whether a journalist opened an email, which is a different event from publishing anything. And the sample is self-selected: it is the customers of one PR software product, sending on the days that product's users send.

PR Gun's 2026 analysis is the largest and the most transparent: 56,606 press releases distributed via ten newswire services, with day-of-week figures drawn from 30,036 dated releases and time-of-day from 22,167. It reports Thursday as the most common issuing day at 22.6%, narrowly ahead of Tuesday at 22.0%.

Read what that measures: when releases were issued, not how they performed. The study is explicit about its own limits — "Because the sample is a 2026 snapshot rather than a multi-year panel, we report distributions within the period and make no claims about year-over-year change." It is a distribution of issuance, honestly labelled. Repeating it as "Thursday is the best day" adds a claim the authors did not make.

Business Wire's press release distribution study analysed over 100,000 releases through its own network. Business Wire sells distribution and does not publish the methodology; we could not read it directly, since businesswire.com sits behind a bot shield that refuses automated requests. Reported, not verified.

CoSchedule's timing guidance rests on a survey of more than 16,000 marketers. That is a large sample of opinions about when to publish, not a measurement of what happened when they did.

Ragan's 2015 study of 100,000 releases is the origin of much of the mid-week advice still in circulation. It is eleven years old, predates the collapse in local newsroom staffing, and predates assistants as a distribution surface entirely.

The circularity problem

Here is the logical difficulty at the centre of all of it. The most-cited datasets measure how many releases go out on each day. If 22.6% of all releases are issued on Thursday, then "Thursday is the best day" is advice to compete in the most crowded slot of the week.

If issuance volume tells you anything about outcomes, it argues for the opposite conclusion — send when the wire is quiet. We are not going to assert that either, because we have no evidence for it. That is the point: nobody has published evidence that settles this in either direction, and the confident advice you are reading is derived from data that cannot support it.

What we could not find

No peer-reviewed study. No study by a party that does not sell press distribution or PR software. No study measuring coverage outcomes — pickup by a named publication, or an article written by a journalist — against send time, with a disclosed methodology and a control. No study that discloses sample composition, sector mix, or the confounders that obviously matter, chief among them that funding and earnings announcements are timed by events rather than by preference and dominate mid-week volume.

If such a study exists, we did not find it in August 2026, and we would rather report the gap than repeat the folklore. The honest version of the answer is: the day of the week is not a variable with a demonstrated effect on press release outcomes, and any vendor telling you otherwise is selling confidence it has not measured.

Four timing inputs that actually matter

None of these is folklore, and all of them are checkable.

1. Regulatory and market timing, if you are a public company. Regulation FD, 17 CFR 243.100, requires public disclosure of material non-public information simultaneously with an intentional selective disclosure, and promptly after an unintentional one. That couples the release to your disclosure obligations and to market hours, not to a marketing calendar. Material announcements are conventionally issued before the open or after the close for reasons that have nothing to do with open rates.

2. Embargoes, if you are also pitching. If a reporter has the story under embargo, the release goes out when the embargo lifts. That is a negotiated time, and it overrides everything else. Distribution and outreach are two different products, and when you run both, outreach sets the clock.

3. Your vendor's cut-off and review queue. Every wire has a submission cut-off, a review step, and a holiday schedule. Self-serve turnaround is typically quoted at 24 to 72 hours, and editorial review can add to it — particularly in restricted categories, where documentation gets requested. If you need a specific publication date, work backwards from the vendor's stated cut-off and confirm it in writing. This is the timing constraint that actually bites, and it is the one nobody blogs about.

4. Your own calendar conflicts. Do not issue against your own earnings, your own conference keynote, or a competitor's known launch date. This is judgement, not data, and it is worth more than the day of the week.

What we tell customers

Send when the announcement is true and approved. If you are listed, send when your disclosure obligations say to. If a reporter has it under embargo, send when the embargo lifts. Otherwise, do not spend a week arguing about Tuesday versus Thursday, because there is no published evidence that the argument has a right answer.

Spend the time on the document instead. The headline, the quote and the structure have observable effects on whether an editorial desk accepts the release and whether a reporter can use it. The send time does not, as far as anyone has demonstrated.

Where we sit

We are a reseller of distribution and we sell 48-hour turnaround, which is a scheduling commitment rather than a claim about outcomes. We do not promise rankings, permanent pages, guaranteed publication, or that any particular send time improves any of them — and we would rather publish the absence of evidence than a chart we cannot stand behind.