Victoria LopianoWhen’s the best time to send an email newsletter?
If we had a dollar for every time we’ve been asked that question, well… let’s just say we’d have a lot of dollars.
We know (and hopefully you do, too!) that the answer to this question can only be determined through testing: testing of your email, against your unique list of recipients, for date, time, and frequency preferences.
So, the better question is: When’s the best time for your company to send your newsletter?
That’s the question we asked ourselves a month ago.
Ready, set, TEST.
It was high time we evaluated our own weekly email newsletter send time, so, about a month ago, we jumped in with both feet to get the firsthand scoop on what our subscribers preferred.
At this point, we had been sending our weekly newsletter at 11 a.m. on Thursdays, and had been doing so for approximately two years.
While I’d like to tell you that we had a solid hypothesis of what would come from our testing, we basically all went into it with limited predictions.
We began by blocking off a three week period where we’d test our newsletter by breaking down our list into six groups and sending on Wednesday, Thursday and Friday at two different times, 11 a.m. and 2 p.m. We chose these times because we wanted to have one morning send and one afternoon send. Note: You could choose to stagger your sends over any time period you’d like, so you don’t have to be limited to choosing just two times. We chose to test for three weeks because we wanted to have conclusive data opposed to testing for a week and possibly getting an outlier that just so happened to occur one day.
The results are in…
We ended the three weeks with a stack of glorious, beautiful analytics. The data nerd inside all of us on the marketing team rejoiced as we poured over the information. What we came to find were statistics that were definitely not what we’d expected, while simultaneously realizing we *did* have some sort of pre-conceived notion of what to expect. For example, I think it’s safe to say we all thought Friday was a dud for sending email, except our analytics now told us otherwise, as outlined below.
The two highest open rates we recorded ended up taking place on a Friday at 2 p.m. and a Wednesday at 2 p.m. These were not consistent across all three weeks, so we chalked it up to a degree of randomness and probably not something that should result in a change (see, this is why we test over three weeks). This is a good statistic to consider in your test, though, and had ours been more consistent, we might have placed more importance on this number.
Moving on to click rate, we didn’t see a distinguished difference from send to send, which led us to factor out that measurement when making our final decision.
And then there’s the click-to-open-rate (CTOR)— which is often considered the most important measurement, as it is the truest indicator of subscriber engagement. Thursday morning sends turned out the highest click-to-open-rates on average, and also boasted our highest CTOR of the entire process.
Now, here is a surprising stat: although the two highest open rates occurred in the afternoon, the click rate and CTOR was significantly lower across the board on afternoon sends when compared to their 11 a.m. morning counterparts.
After reviewing all of this information, we had a decision to make. Although open rates were higher on the two days that weren’t Thursday, the most important metric, CTOR, was consistently higher on Thursday mornings.
So, do we entertain sending on one of the days with a higher open rate and assume the CTOR might naturally increase over time? Or do we stick to what we know has higher engagement?
At the end of the day…
We ended up sticking to what we know. We presume the reason the CTOR is so high on Thursdays is because we’ve conditioned our dedicated readers to expect our newsletter on Thursday at 11 a.m., so now they look for it at that time (maybe even subconsciously).
Did we make the right choice? You tell us! What would you have done? Have you performed tests on your newsletter recently?
(192)
