Published best times are averages across unrelated industries. Test your own list over several sends, and expect timing to matter less than the sender, the subject, and the content.

Where the advice comes from

Studies of send times analyse enormous volumes of messages across every kind of sender, and report an average.

The averaging is the problem. A business emailing office workers, a shop emailing consumers, and a trade supplier emailing site managers have different audiences with opposite rhythms.

An average across all of them describes none of them, and the specific hour it produces is an artefact of the mix rather than a finding.

Which is why the recommended day changes every few years and why every study finds something slightly different.

The reasoning against yourself

Worth noticing, because it defeats the advice on its own terms.

If Tuesday at ten is genuinely the best time, then everybody following that advice sends at Tuesday at ten, which makes it the most crowded hour in the inbox.

A message arriving at a popular time competes with more messages, and a message arriving when the inbox is quiet is visible for longer.

So widely followed advice about timing tends to destroy the advantage it describes.

That is not an argument for any particular alternative hour. It is a reason to treat the published figure as a starting guess rather than as an answer.

What actually varies by audience

Those are more useful than an hour, because they describe when your particular readers are likely to be looking rather than when the average of everybody is.

Test your own list properly

The method, and the main constraint is patience rather than difficulty.

Pick two candidate times drawn from what you know about your customers, and alternate between them for several months.

Compare across at least five or six sends rather than one, because on a small list a single comparison is noise.

Keep everything else the same while you test: same day of the week if testing time, same sender, similar subject style.

Record each result in the same place, since the point is a pattern across months rather than a result from one send.

Where the two times produce results within a few percent of each other across six sends, the honest conclusion is that it does not matter for your list, which is a genuinely useful finding.

A worked example

A business serving trades customers sent their monthly message at ten on Tuesdays, following the standard advice.

Opens were reasonable and replies were rare.

Somebody pointed out that their customers were on sites at ten on Tuesdays and read email at either six in the morning or after seven at night.

They alternated between the two for six months.

Early morning outperformed the mid-morning slot consistently, and evening was similar to it.

The improvement was real and modest, roughly a fifth more opens, and considerably smaller than the improvement they had got earlier from changing their subject lines.

Their conclusion was that timing was worth getting roughly right and not worth optimising further.

Consistency beats optimisation

The finding that matters more than the hour.

A message arriving at the same time each month becomes expected, and expected messages get opened by people who recognise the pattern.

That effect is larger on a small list than any timing advantage, because a recognised sender arriving at a familiar moment is close to correspondence.

Which argues for choosing a reasonable time and keeping it, rather than moving it in pursuit of a better one.

The same applies to frequency: monthly, reliably, outperforms an irregular schedule averaging the same volume.

Time zones and delivery timing

Two practical points that are more concrete than the folklore.

If your list spans provinces, the send hour differs for each recipient, and a tool that can send by local time is worth using if you have one.

For a business serving one area this does not arise, which is most of the audience for this advice.

Also worth knowing: sending is not instantaneous, and a large send from a modest service can take a while to work through, so the scheduled time is when it starts rather than when it arrives.

For a few hundred addresses that is minutes rather than hours, and it is worth knowing before you attribute a result to an exact hour.

What to test instead

Since testing capacity on a small list is limited and should go where the returns are.

Subject lines, which produce larger differences than timing and can be tested within a single send.

The sender name, which affects the decision more than either.

Frequency, which is the change most likely to alter results substantially on a small list.

And what the message actually contains, which is the largest factor and the hardest to test.

Timing belongs somewhere below all of those, which is roughly the opposite of the attention it receives.

The day matters more than the hour

A distinction worth drawing, since the two get tested together and behave differently.

The hour affects where in a stack of messages yours sits, which matters for a few hours and then stops mattering.

The day affects whether somebody is working, on site, at home, or away, which changes their whole relationship to their inbox.

So a business whose customers do not work Mondays has a real reason to avoid Mondays, and that is a larger effect than any choice of hour within a working day.

Test the day first, across a few months, and treat the hour as a refinement afterwards if the day turns out to matter at all.

The counter-case

Timing is not entirely folklore.

Some send times are genuinely poor: late Friday afternoon, the middle of the night, and public holidays all reliably underperform for most audiences.

Anything time-sensitive obviously has to arrive in time to be acted on, which is a constraint rather than an optimisation.

And for a business whose customers have a very distinct rhythm, such as one serving restaurants or schools, the timing question has a real answer that is worth finding.

Avoid the obviously bad slots, pick something plausible for your customers, keep it consistent, and spend the testing effort on the subject line.

What to do

  1. Ignore the published best hour.
  2. Think about your customers' day instead.
  3. Pick two plausible times.
  4. Alternate for six months, recording each.
  5. Accept no difference as a finding.
  6. Then keep it consistent.
  7. Spend testing effort on subject lines instead.

Step five is the outcome most small businesses reach, and knowing that timing does not matter for your list frees the attention for something that does.

Why single results mislead at low volume is covered in measuring a site with fifty visitors a day.

Frequently asked questions

Where does send-time advice come from?

Averages across enormous volumes of messages from every kind of sender. The averaging is the problem, since audiences with opposite rhythms are combined into one figure.

Does following the advice help?

It partly defeats itself. If everybody sends at the recommended hour, that becomes the most crowded time in the inbox and a message is visible for less time.

What varies by audience?

Trades and site workers read early or in the evening, office workers mid-morning midweek, consumers at home in the evening. That is more useful than a specific hour.

How should I test?

Alternate two plausible times across five or six sends, keeping everything else the same, and record each result. A single comparison on a small list is noise.

What matters more than timing?

The sender name, the subject line, frequency, and what the message contains. Timing belongs below all of those, which is the opposite of the attention it gets.

Is any timing genuinely bad?

Yes. Late Friday afternoon, the middle of the night, and public holidays reliably underperform. Avoid those, pick something plausible, and keep it consistent.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Sending at ten on Tuesday because you read it somewhere?

Think about when your customers are actually holding a phone. For trades that is rarely mid-morning.

Start a Conversation