By David Smith
As an email marketer, you want the best open rates you can get. A big part of that is sending your emails at the right time and on the right days. But what are those best times and days?
If you Google “Best time and day to send marketing emails”, you’ll end up with about a million articles (4,150,000 to be exact), all purporting to know best. Some of these claims are backed by data, some are backed by “years of marketing experience”, and some just come out of the blue.
If you want to be a generic marketer, you can certainly follow these generic articles’ advice. But if you want to be a hugely successful marketer, with answers that are uniquely catered to you and your email list, you’re going to have to work a bit harder.
In short, you’re going to have to experiment.
I’m not impressed, and as an engineer, I’m also not one to take advice without doing my own research first, using a little thing known as the scientific method. Stay with me. We’ve already accomplished the first step by establishing a question: When should we send marketing emails in order to get the best open rates?
Next, let’s look at the claims that are floating around out there today, both to help us form a hypothesis, and to understand what others are saying already. A cursory Google search brought me to a popular article claiming to have aggregated data from 10 different sources, telling us that sending emails on Tuesday at 10 a.m. gives you the best open rate. Easy—just send your emails at that time, and they’ll get the best visibility, right? Sure, along with all of the email from your competitors and everyone else vying for your audience’s attention. Following general advice like this also assumes that your audience is just that: generic.
While I’m not going to blindly follow that advice, it does give us a good starting point to test against: “Sending an email at 10am on Tuesday gives us the best open rates.” It also stands to reason that this is a safe starting point to test against. Unless you have prior data suggesting otherwise, following this advice isn’t likely to tank your marketing campaign, so we can experiment with minimal risk.
Infusionsoft (or any other marketing platform worth using) allows you to send email broadcasts and have them delivered at the correct time regardless of their time zone and then aggregates the open and click rates, so we’re able to test our hypothesis and gather meaningful data. How meaningful will depend on the number of contacts you’re sending your email to, with more being better, but as long as your customer base stays relatively constant during your experimentation, we should be fine.
Now we need to begin actually conducting our experiments. Ideally, you have a massive contact database with hundreds of thousands of potential leads, and you can conduct very large experiments with large groups of people receiving identical content on different days, right? Ok, probably not, especially if you’re a small business, so the important thing here is to limit the number of variables and gather as many data points as you can.
Your brand probably has a voice already, and you probably have a marketing style. Try to be consistent as you conduct your experiment. If you send out one email with a short, to-the-point subject line, and a second with a lengthy ambiguous subject line, you’re potentially skewing your results.
Unless you do actually have that massive customer database, where you can segment and send emails to groups of tens of thousands of people, keep your audience consistent while you experiment. If you send out an email on Tuesday and get a 27 percent open rate, but next week you send out an email on Wednesday and get a 20 percent open rate, be sure you didn’t do something to significantly increase, decrease, or modify the contacts you’re sending to. New leads are fine, but if you run a major lead-capture campaign during your trials, you may need to discard your data and start over.
Finally, establish as many data points as you can. If I send an email on Tuesday and receive a 15 percent open rate, an email on Wednesday with a 20 percent open rate, and an email on Friday with a 12 percent open rate, what have I figured out? That Wednesdays are the best day for me to send emails? Absolutely not. One email is insufficient to make a determination. Ideally, send a dozen emails at the same time on the same day of the week, and aggregate that data in order to determine what the real result of your trial is.
In the end, is all of this experimentation and hypothesis testing more difficult and time-consuming than reading an article online? Heck yeah. But it also gets you better data—and data that’s specifically applicable to you and your audience.
Sure, try sending all of your emails on Tuesday at 10 a.m., but if you’re selling a hangover remedy, your audience may be a little more receptive at 3 a.m. on a Saturday or 2 p.m. the next day. If you’re advertising vacations, 9 a.m. on a Monday may be a better time to tap into your customer's desires.
The best way to find out? Experiment.
David Smith is a Manager of Software Engineering at Infusionsoft.