Not legal advice: This article covers marketing practice and general legal context, not legal advice. Messaging law (like the U.S. TCPA) changes over time and applies differently case by case — verify your specific obligations with a qualified lawyer before you rely on anything here.
Search "best time to send an email" and you'll get the same answer everywhere: 10 a.m. on Tuesday (sometimes Thursday), with Monday and Sunday flagged as mistakes. It's repeated with total confidence across hundreds of marketing blogs, usually next to a confident-looking bar chart and a citation to "studies" or "billions of emails analyzed." WhatsApp gets its own folk version — send between 8 and 10 a.m., or during the lunch hour, or in the evening, depending on which vendor blog you land on first.
We went and checked the sources those charts actually point to: two competing platforms' real, disclosed datasets covering more than 30 billion sends combined, the one peer-reviewed study we could find that tried to model this rigorously, and WhatsApp's own published rules for when a business is even allowed to message someone. What we found: the studies behind the folklore disagree with each other, one of them quietly undercuts the exact claim it's cited to support, and the one rigorous academic attempt at this problem concluded the entire premise — that there's a single shared best hour — is close to backwards.
Key takeaways
- The "Thursday at 10am" recommendation cited across marketing blogs mostly traces to one aggregator — CoSchedule's "What 10 Studies Say" post — which vote-counts 10 vendor blog posts (2020-2022, no disclosed sample sizes) rather than pooling real data. Thursday "won" by ranking #1 in 4 of the 10 posts, not by beating other days in a shared dataset.
- Omnisend's own disclosed 2026 dataset — roughly 26 billion real emails, one of the very sources industry roundups cite — shows day-of-week open rates clustered within 1.6 percentage points of each other (29.67%-31.27%). Sunday, the day most roundups call the worst, produced the single strongest hour in the whole dataset: an 11 a.m. open rate of 33.69%.
- GetResponse's disclosed 2023 dataset — 4.4 billion messages — reaches a similar conclusion from a different angle: weekday differences are "little," while the hour right after send matters far more, capturing 21.2% of all opens and 44.14% of all clicks.
- The one peer-reviewed attempt we found to model this rigorously (Applied Sciences, 2022) concluded the accurate answer isn't a shared calendar slot at all — it's a per-subscriber prediction built from that person's own open history, reaching a 0.91 R² score when modeled individually.
- No comparable disclosed research exists for WhatsApp send times — every "best hour" claim we traced led to an uncited marketing blog, not a study. What is real and Meta-enforced instead is a 24-hour customer-service window: free-form replies only within 24 hours of a user's last message, a pre-approved template required outside it.
- The U.S. legally mandates SMS and call quiet hours (the TCPA: no calls or texts before 8 a.m. or after 9 p.m. local time, $500-$1,500 in statutory damages per violation) but has no equivalent time-of-day law for email or WhatsApp — and Israel's own workweek (Sunday-Thursday) makes "avoid Sunday" advice describe the wrong day entirely for an Israeli business.
What is the "10am Tuesday" rule, and where does it actually come from?
The claim shows up in near-identical form across thousands of marketing posts: a specific day (usually Tuesday or Thursday) and a specific hour (almost always 10 a.m.), presented as the scientifically optimal send time, next to a bar chart and a vague citation to "studies" or "data."
One page keeps surfacing as the likely source most of that folklore traces back to: CoSchedule's "What 10 Studies Say About the Best Time to Send Email." It's exactly what the title says — a roundup of 10 separate vendor blog posts (Sendinblue, GetResponse, Omnisend, HubSpot, Salesforce Marketing Cloud, Moosend, Campaign Monitor, Sender, MailerLite and Mailjet, published 2020-2022) with a specific counting method: it tallies how many of the 10 posts rank each day or hour #1, then declares the day with the most #1 votes the winner. None of the 10 source posts discloses a sample size in the roundup itself.
| What the roundup claims | What we found checking it |
|---|---|
| "Backed by 10 studies" analyzing billions of emails | The 10 sources are vendor blog posts spanning 2020-2022 — not one discloses a sample size, date range or methodology in the roundup that cites them. |
| Thursday is the best day, chosen from real data | Thursday is a vote count: it ranked #1 in 4 of the 10 posts and #2 in one more. That's popularity across 10 separate opinions, not a measurement across one shared dataset. |
| Sunday is the worst day to send | One of the very studies the roundup cites — Omnisend's own dataset, below — shows Sunday landing within 0.7 points of the week's average open rate, with the single best hour of the entire week falling on a Sunday. |
None of this makes the 10 source studies fake — GetResponse and Omnisend are real platforms publishing real, if platform-specific, numbers. What's manufactured is the certainty: a vote count across 10 differently-run posts, most with no disclosed sample size, presented as if it were one precise, science-backed answer.
What does real, disclosed send-time data actually show?
Omnisend, an email marketing platform, published its own 2026 research analyzing roughly 26 billion emails sent by its customers — real scale, a disclosed source, and it's one of the studies industry roundups cite. Its day-of-week open rates are strikingly flat: every day falls between 29.67% and 31.27%, a spread of just 1.6 percentage points across the entire week. Tuesday narrowly leads on opens (31.27%) and clicks, but Friday leads on conversions (0.081%) despite not being the best day for opens at all — and Saturday is the weakest day for clicks and conversions and the second-weakest for opens (29.99%, just above Monday's 29.67%).
The report's most interesting single number cuts directly against the "Sunday is a mistake" folklore: at 11 a.m., Sunday produced a 33.69% open rate — one of the strongest single hours anywhere in the dataset, on the exact day most roundups tell you to avoid.
Omnisend's own report draws the same conclusion this data implies: the gap between the best and worst day is smaller than what a single subject-line A/B test typically produces, so the day should usually be picked for its click or conversion pattern — not chased for opens alone.
Is there a golden hour within the day?
GetResponse, a competing platform, analyzed 4.4 billion messages sent by its own customers in 2023 (accounts with at least 500 contacts) for its Email Marketing Benchmarks report. Its day-of-week finding echoes Omnisend's: "there's little difference in terms of engagement between individual weekdays" — Tuesday edges ahead on average, but not by a margin worth building a schedule around.
Where GetResponse's data does show a real pattern is timing after the send, not the calendar slot itself: the first hour captures 21.2% of all opens and 44.14% of all clicks, and roughly 70% of opens and 85% of clicks happen within the first 24 hours. The report also flags two windows — very early morning (4-6 a.m.) and early evening (5-7 p.m.) — as typically less cluttered inboxes, though it's explicit that this matters far less than the audience itself.
21.2%
the share of all email opens that happen in the first hour after send, per GetResponse's analysis of 4.4 billion messages sent by its customers in 2023 — the clearest real timing signal in the data, and it has nothing to do with which day or hour you picked.
Source: GetResponse, "Email Marketing Benchmarks" (4.4B messages, 2023)
GetResponse's own report states the underlying point plainly: "If you put your audience first, they'll engage with your emails no matter what day you send them."
Is there real science behind "send-time optimization" — or is it marketing?
One genuinely rigorous academic attempt at this problem does exist. Carolina Araújo, Christophe Soares, Ivo Pereira, Duarte Coelho, Miguel Ângelo Rebelo and Ana Maria Madureira published "A Novel Approach for Send Time Prediction on Email Marketing" in the peer-reviewed journal Applied Sciences (MDPI) in 2022. Instead of asking "what's the best universal hour," they built models that predict the best send time for each individual subscriber, based on that person's own history of opens, clicks and interaction timing — testing several regression algorithms (Random Forest, Multiple Linear Regression, K-Neighbors, Support Vector Regression) and combining them into a stacking ensemble.
The individualized, per-subscriber model was the point, and it worked: the best-performing stacking ensemble reached a 0.91 R² score with a mean absolute error of 0.204 — accurate enough to predict a given subscriber's best send day to within roughly half a day. That's a real, measured result for personalization. It's also the opposite premise from "everyone should get emailed at 10am Tuesday": the paper's whole approach assumes the best time is a property of the individual recipient, not a shared calendar slot at all — which is exactly why the disclosed platform data above is so flat when averaged across millions of different people with different habits.
Does automation beat guessing the perfect hour?
If the real signal is individual behavior rather than a shared clock, the practical version of that insight isn't a fancier prediction model — it's sending based on what a specific contact just did, not on a scheduled broadcast at all. Klaviyo's own SMS benchmark data shows exactly how much that distinction is worth: automated, trigger-based "flow" messages (sent because a contact did something — abandoned a cart, made a purchase, missed an appointment) account for just 7.6% of all SMS sends on its platform, but drive 45.2% of total SMS revenue. Their average click-through rate (nearing 10%, with top performers above 16%) also runs close to double what broadcast campaign sends typically achieve.
Share of sends
Share of revenue
In oToK, this is what an Automation's trigger does: a flow can start the moment a real event happens to a specific contact — a sale logged, an event RSVP missed, a cart left behind — rather than on a calendar guess. The Wait step inside a flow adds timezone support and its own wait rules, so a message can still land at a sensible local hour for that one contact without needing a universal "best time" for everyone on the list.
Does WhatsApp have its own best hour — and did we find real data for it?
We looked for a WhatsApp equivalent of Omnisend's or GetResponse's disclosed datasets — a real platform, a real send volume, a real measured open-rate-by-hour breakdown. We couldn't find one. Every "best time to send WhatsApp messages" article we checked converged on the same generic windows — morning (8-10 a.m.), lunch (12-1:30 p.m.), evening (6-8 p.m.) — without naming a platform, a sample size, or a study behind the number. It reads like the email folklore one step further removed from any disclosed source at all.
What is real, documented and directly enforced by Meta is a different kind of timing rule — not about the best hour, but about whether a business can message someone freely at all. A conversation opens a 24-hour customer service window from a user's last inbound message: inside that window, a business can send free-form replies. Outside it, only a pre-approved Message Template — reviewed by Meta in one of three categories (Utility, Marketing, Authentication) — can be sent, and approval commonly takes minutes to about 24 hours.
| Inside the 24-hour window | Outside the 24-hour window |
|---|---|
| Free-form message, any content, any time | Only a pre-approved Message Template |
| No Meta review needed | Meta reviews each template (minutes to ~24 hours) |
oToK's WhatsApp Campaigns feature is built directly around this rule: templates sync two-way with Meta, with approval status (and rejection reason, if declined) visible on the Templates page before you schedule a send — so the actual constraint that governs when a message can go out is handled before send time, not guessed at with a golden hour.
Are there real quiet-hours laws you need to follow?
Unlike "best hour" folklore, this one is genuinely legal, not a marketing convention. The U.S. Telephone Consumer Protection Act (TCPA), as implemented by FCC regulations, prohibits telephone solicitation calls and texts to a residential number before 8 a.m. or after 9 p.m. local time at the called party's location — and violations carry statutory damages of $500 to $1,500 per message, which has made TCPA quiet-hours claims a real and active category of class-action litigation against consumer brands. Some states go further: Texas Senate Bill 140, in force since September 1, 2025, adds its own quiet-hours window for text solicitations to Texas residents — 9 a.m. to 9 p.m. Monday through Saturday, and just noon to 9 p.m. on Sundays and holidays, starting later than the federal TCPA window every day of the week.
Notably, platforms often recommend being more conservative than the law strictly requires — Klaviyo's own guidance to its SMS customers is to avoid sending before 9 a.m. or after 8 p.m., a full hour tighter on each end than the federal minimum. Email carries no equivalent U.S. federal time-of-day statute, and WhatsApp's Meta-enforced rule (above) governs the 24-hour conversation window rather than a specific clock — so of the three channels, SMS and voice are the ones with an actual, enforceable send-time law attached to them.
So what should you actually do about send time?
- 1Stop optimizing for a universal hour. The two largest disclosed datasets we found (Omnisend, GetResponse) both show day-of-week differences small enough to be noise — spend that effort on the subject line, the offer or the segment instead.
- 2Pick the day by the metric you actually care about, not opens. Omnisend's own data shows Friday out-converting Tuesday despite a lower open rate — decide what "best" means for your campaign before you pick a day for it.
- 3Respect the timing rules that are real, not folkloric: U.S. SMS and calls have a legally enforced 8 a.m.-9 p.m. quiet-hours window; WhatsApp's real constraint is the 24-hour customer-service window and template approval, not a golden hour.
- 4Wherever it's practical, send on the trigger, not the clock. A message tied to what a specific contact just did (a purchase, a missed appointment, an abandoned cart) reaches them at a meaningfully better moment than any shared calendar slot could — Klaviyo's own flows-vs-campaigns split above shows the size of that gap.
- 5If you do send Israel-specific broadcasts, don't import Western "avoid Sunday" advice as-is — Sunday is the first business day of Israel's Sunday-Thursday workweek, not the tail end of a weekend, so the day it's describing simply isn't the day being sent to.
In oToK, a WhatsApp campaign's Step 2 lets you pick the timing per send — now or scheduled for a specific date and time — and every campaign runs through a managed send queue with automatic retries, so a large list still goes out at a safe, steady pace rather than all at once. For anything trigger-based rather than a broadcast, an Automation's Wait step adds timezone support and its own wait rules, so a flow can land at a sensible local hour for each contact without hand-picking one send time for the whole list.
Frequently asked questions
- Is 10am on Tuesday really the best time to send a marketing email?
- Not based on any disclosed dataset we could find. The "Thursday/Tuesday at 10am" claim traces mainly to a roundup that vote-counts 10 uncited vendor blog posts rather than pooling real data. The two large, disclosed datasets we checked (Omnisend, ~26B emails; GetResponse, 4.4B messages) both show day-of-week differences small enough to be noise — well under 2 percentage points across the whole week.
- Does send time matter at all, or is it a distraction?
- It matters less than most guides claim, but not zero. GetResponse's data shows the hour right after you send captures a disproportionate share of opens and clicks (21.2% of opens in hour one), and real, legally enforced quiet-hours rules exist for SMS and calls in the U.S. What doesn't hold up is the idea of one universal "best" day or hour for everyone on a list.
- What's the best time to send a WhatsApp message?
- We couldn't find a single disclosed, measured study behind any of the "best WhatsApp hour" claims circulating online — they read as folklore, not research. What is real and enforced by Meta is the 24-hour customer-service window: free-form replies are only allowed within 24 hours of a user's last message, and a pre-approved template is required outside it. That constraint matters far more than any guessed hour.
- Should an Israeli business follow the "avoid Sunday" advice from U.S. studies?
- No — that advice describes the U.S. weekend, and it doesn't translate. Israel's standard workweek runs Sunday through Thursday, with the weekend on Friday and Saturday, so Sunday is the first business day, not a low-engagement tail end of the week. If anything, the U.S. data itself argues against blanket "avoid Sunday" advice even on its own terms — Omnisend's dataset found Sunday's best hour outperformed most weekday hours.
