When is the Best Time to Send?

If I had a nickle for every time someone asked me this… You can see the results of a survey we did on this a little while back. However, as you’ll see if you read that, it depends for everyone. So let’s look at how you can find out which time is best for you.

The Reporting tab in Bronto is a great starting place for digging into the information you already have. In this tab we can dig through the gold mine of historic sending data Bronto has stored for you. To start things of lets look at the Message Performance After Send chart.

Here we are seeing the opens for a specific campaign in terms of how many hours after the initial send they occurred (note that each bar represents a three hour time frame). Before we dig into what this reveals, I should explain that there is usually a spike immediately after you send a message.

Now this chart is very telling in several ways, let’s examine them:

• There is not just a spike after the send, there is a massive spike. And not only is the spike large, it dies down pretty quickly.
• The spike also re-occurs every 24 hours for the next two days. That speaks volumes. We’ve nailed the peak time, even days later people are coming back at that same time to read the message.

If we dig in and look at Message Performance by Hour, then we will be able to see a very tight cluster around the mid-day mark. Again, this is skewed by the fact that we send the message around then. However, the fact that we don’t see any outlier spikes tells us that this time is working for most people.

Now the next question is what day of the week should we be sending. That’s a harder one to tell. If the other two graphs were slightly skewed by our own sending paterns, then this one is beyond biased. If you always send on Tuesdays (we usually do in this case), then you are definitely going to see a spike there.

I’m encouraged here by the fact that we are seeing a very sharp trail off on the days following the send. I’d always expect to see some next-day opens.

Let’s give ourselves a pat on the back, we’ve done a great job here – WAIT! You can’t quit that quickly. With all of these charts I had to call out the fact that our data was skewed because we are following best practices with relatively consistent date/time deliveries. The only way to know that we are right is to test this out.

That’s right, once you have looked at the data you should consider running some basic tests. From the data we have seen here I’m going to try out sending on Wednesdays at 2:00PM rather than our usual Tuesday at noon. Using the easy Date/Time Champion/Challenger tests (which we are releasing in just a few weeks!) from within Bronto this will only take me an extra minute or two to set up when I do my next delivery.

It’s as easy as that, within a few minutes we’ve looked through the data and feel pretty good about where we are. We’ve also figured out a plan for a simple test to see if we really are hitting the optimal date and time for our list. I should have results of this test in the near future. I’ll let you know how it goes.

Try this out with your own site, let me know what you find. If you see something puzzling, interesting, or downright amazing then please leave a comment, I’d love to hear about it!

Adam Covati
Product Manager at Bronto

Related posts:

1. Making The Grade With Reporting Remember the days when Mom and Dad used to hound...
2. New Feature Alert – Segmenting on Relative Dates Have you ever wanted to create a segment in your...

Tagged as: Message Performance, Message Performance After Send, Message Performance By Hour, Reporting->Charts, Reporting->Dashboard, Time of Day