orginaly published by Silas Partners September 2006

The most frequent email related question we hear at Silas Partners is: “When should we send out our mailing?” The industry standard reply is by and large: early in the week, early in the day. This conclusion is often based on open rate analysis.

I’m seldom happy with “industry standard” and “by and large.” In order to be able to give more specific and more helpful advice, you have to go past open rates when looking at email performance. To this end we recently compiled a sample of email statistics and performed some new analysis on this data. The results validate some of the industry standard thought, but they also give more insight.

We applied a new analysis to the open, click through and action rates of the emails we analyzed. An action is defined by a user filling out survey, making a donation or taking an advocacy action. Normally mail performance is judged on one of these factors alone, but this narrow view doesn’t take the whole picture into account. You need to look at how these factors influence each other to get a more complete picture of the success of an email campaign.

An example – with small numbers – will make this more clear. Say you have a list that has ten subscribers, and you send out two emails to this list. The first one is opened by five people but only one clicks through to take action. The second email is opened by three people but two people click through to take action. Which email was more effective? If one applied simple open rate analysis the second email would have appeared to fail when in reality it faired much better at eliciting action.

Often time these numbers are looked at without considering the percentage difference from the average. Without looking at the percentage difference it is difficult to decide if any increase is actually significant. Another example will help to illustrate: If you have an open rate on Monday of 15 percent and 20 percent on Tuesday normal analysis would say you should send emails out on Tuesday. However, if your average open rate is 17 percent then it is evident that there isn’t a truly significant difference between Monday and Tuesday.

Enough with the theory, now on to the numbers. First, a quick note about our sample; we used a cross section of types of email and frequency. We also excluded daily or weekly devotional emails from our analysis as these tend to be more about keeping in touch and providing a service and less about eliciting action. While our sample is representative it is on the small side by statistical standards, which means that your results may vary - more on that later.

From our analysis the best time to send an email is Tuesday between 3pm and 6pm. We would not have arrived at this conclusion had we used traditional open rate analysis. Tuesday actually has the second to worst open rate of any day of the week. However, the click through, action, and forward rate are by far the best of any day of the week.

The general concept of our analysis was to get a number of opens, clicks, actions, and forwards you could expect per 1,000 emails sent. On average our data showed one could expect 243 of those 1,000 emails to be opened and of those 243 opens we expect 38 clicks, 24 actions and 0.6 forwards.

On Tuesday you could expect 211 opens, 73 clicks, 51 actions and 1.1 forwards. That is 91 percent more clicks, 117 percent more actions and 70 percent more forwards.

If we had applied simple open rate analysis to our data set we would have suggested Thursday as the best day to send emails. Our data showed one could expect 300 opens for every 1,000 emails sent.

However, the click, action and forward rates plummet on Thursday. All three rates are down between 60 and 70 percent leading to significantly fewer actions per 1,000 emails sent.

This clearly points out the problem with the question, “What is the best day to send our emails?” The objective of the email weighs heavily on the answer to that question.

Some emails are focused on getting information to constituents. If that is the goal, Thursday, with the highest open rate, is your best day. However, if you are looking for the best overall performance, Tuesday, with the highest action rate, is the obvious choice.

We also found that time of day can have an impact on the performance of your emails, although the variations were less than the variations caused by the day of the week. In general we found that sending messages overnight would give you the best action and forward performance, while sending them after 6 pm eastern would give you the highest open and click rate.

When we combined time of day for Tuesday we found that Tuesday between 3 pm and 6 pm (EST) had the highest action and forward rate. It is important to note that as with any set of data, the more you slice it the smaller your sample becomes and the more prone to variation and error any analysis becomes.

Also to reiterate, while we feel confident in our analysis, our sample size is not large enough to make broad general claims about every email you would send. The original purpose of the analysis was to collect data and look at open rates to make such a general claim. However, through the analysis we have found that while claims in the aggregate may prove instructive, there are more nuanced factors that have to be considered.

To summarize, the bad news is: knowing when to send your emails isn’t as simple as you thought it was. The even worse news is that depending on your audience and your email content our aggregated analysis may not match up to what you are seeing. The good news is that Silas Partners is ready to help.