Best time to send: Open rate decay (part 2)

Here’s one of the most popular questions we’re asked all the time:

When is the best time to email?

There’s really only one answer to this question: test. Test and measure rigorously for when the best time to email is. Let’s explore how we did our own testing internally at WhatCounts to determine when the best time to send an email is. We begin with our standard measures of email performance: open rate, clickthrough rate, and conversion rate.

A couple of months back, we asked the question about whether moving from a Thursday mailing to a Tuesday mailing would be more impactful to our business based on a hypothesis about Pareto curves. Our testing scenario was to see whether moving to Tuesdays would generate more opens. The answer?

Measuring open rate decay is important.

It turns out that Tuesday openings get a strong initial pop – that is, people are more likely to open them on a Tuesday than a Thursday. But our initial hypothesis about the Pareto curve was incorrect; instead of a sustained higher open rate, once Tuesday ended, the impact of the mailing diminished much more rapidly than a Thursday mailing. The Pareto curve got flatter, which means that fewer and fewer people were opening the email.

Open rate is the top of the email marketing action funnel. Let’s take a step down the funnel to clickthrough rate. On average, how did the mailings perform?

Tuesdays: 0.94% clickthrough rate
Thursdays: 0.86% clickthrough rate

Now we’ve got some interesting conflicting data. Thursdays seem to be better for open rates, but Tuesdays seem to be better for clickthrough rates. Let’s take a final step down the funnel to conversion to a lead, which is our primary goal as marketers.

Tuesdays: 42.75 leads generated on average
Thursdays: 67 leads generated on average

There’s a very sharp difference there. Any salesperson would gladly have more leads than fewer, which is the primary output of an inbound marketing program. For the metric we care about most, Thursday is the best day for our list to be mailed.

If we dig deeply into our conversion analytics and do a custom report on goal completions by source, using the WhatCounts platform as the source, and doing a day of week dimension, we reveal something quite surprising:

Looking at conversion analytics of the newsletter.

Even when we send newsletters on Tuesdays, the highest converting day of the week for our specific list is Thursday. Thus, if we send our email newsletter on the day of the week when the list is most ready to convert, we align the initial open rate burst with conditions most favorable to conversion and get the maximum impact from our list.

There are three takeaways from this test that you should implement in your own email marketing program:

1. Measure by the goal that matters most to you. In our initial examination, we looked at open rate as our hypothesis, but when we dug deeply into our data for the metric that mattered most to us, we arrived at a conclusion that will have a real business impact.

2. Test with an open mind. We still haven’t speculated as to why Thursdays are better for our list than Tuesdays for conversion; in some ways, it really doesn’t matter all that much as long as we know what we should be doing. If we went in trying to prove an opinion or conclusion, we would have looked at our data with a skewed perspective, attempting to justify our position.

3. Give yourself time to test. It would have been tempting to rush out immediately after one or two mailings with a conclusion about what was working and what wasn’t. However, one or two data points don’t make for a statistically significant conclusion.

Finally, this blog post also serves as an announcement that we’re moving our weekly newsletter back to Thursdays for what should be obvious reasons. We hope that you can take the methodology and logic we’ve applied to our own newsletter and apply it to your email marketing efforts for maximum benefit.

Christopher S. Penn

Share on LinkedIn10Tweet about this on Twitter17Share on Facebook0Share on Google+4

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>