When Good Enough Shouldn’t Be: The Best Time to Send Emails

When is the best time to send email? This question has been one of the most asked in the email marketing industry for well over a decade. And the urgency to answer this question correctly has only grown.

That’s because we live in a time-strapped world where we only want things when we want them. But when we're ready, we want them immediately. In the meantime, email messages are coming in all day, every day—largely without regard to when customers are most interested in reading and responding to them. It’s a mismatch that leads to frustration, disengagement, and opt-outs.

Thankfully, this doesn’t need to be the case. Brands have more data and tools at their disposal than ever to answer the question of the best time to send emails. Patterns in historical open, click, and conversion behaviors are readily available for your each of your subscribers. Taking advantage of that known behavior offers a competitive advantage that will boost performance of your program. In all of my years of working with clients, I've yet to see it fail to deliver material lifts in a program. Ever. 

In fact, among the brands that Oracle Marketing Cloud Consulting has worked with, optimizing the best time to send emails results in an average lift in engagement of 10%-12% and an average lift in conversions of 5%-7%.

However, there are several ways to determine the “best time to send emails." They vary in terms of the level of impact that they'll have, but they're all better than an arbitrary deployment time for your emails based solely upon hunches or how you’ve done it historically. Let’s look at a few data-based approaches to picking deployment times:

A Good Time to Send Emails

Keep it simple: You can get an advantage in the inbox by doing an aggregate analysis of your subscriber base and just examining the timing opportunity by daypart. Look at your send time and then compare it to the percentage of your subscribers opening in the morning, afternoon, and evening.

We recently examined a telco client's data and found that while they deployed their weekly email in the morning, the majority of engagement was actually in the afternoon. We ran an A/B test and found that by shifting the full send to the afternoon, conversions increased 18% when compared to a morning send control group. Easy change, big impact.

The Better Time to Send Emails

You get from Good to Better by going from the best time to send based on aggregate data to send times based on individual-level data. By using data on the timing of past opens, clicks, and conversions for each subscriber, you can determine the optimal time of day at the individual level by day of week. That means that there's a bespoke value for each subscriber for Monday, Tuesday, and the other days of the week.

For example, your subscriber Steve should receive your emails at 7am on Mondays, 11am on Tuesdays, and so on; whereas, Tara should get them at 10 am on Mondays, 3pm on Tuesdays, and so on.

At the very least, there’s a best time to send emails for each person regardless of day. For instance, Steve should get email at 1pm and Tara should get email at 4pm, whether you’re sending on Monday or Saturday.

Oracle Responsys and Bronto each support send time optimization. Oracle Eloqua currently offers it as a Controlled Availability program. 

The Best Time to Send Emails

The best option is a combination of the two approaches above. Why? Because you won’t have sufficient data to determine an optimal send time for a significant portion of your list - perhaps up to 30% of it.

New subscribers will make up the vast majority of this group. With no pattern of engaging with your emails, it’s impossible to determine an ideal individualized send time. In these cases, rather than setting a default send time arbitrarily, use your aggregate best time to send emails as the fallback. For everyone else, use their bespoke optimal send time.

The Winning Metric

When determining the best time to send emails, we recommend using a metric that’s aligned with the goal of your message—which is generally clicks, if not conversions. Even if you’ve written a clear and descriptive subject line and preview text that’s designed to drive bottom-of-the-funnel activity, time of day can affect follow-through based on the type of call-to-action.

For instance, you might find that subscribers are more likely to act on call-to-action to share content via their social media accounts in the morning, more likely to register for a webinar or event around lunchtime, and more likely to make a consumer purchase in the evening. Keep this in mind.

Caveats & Exceptions

Note that there are times when you might want to override send time optimization, or you might want to put some specific parameters in place regarding windows during which the optimization occurs. For example:

• Store hours. If your emails tend to drive traffic to your stores or restaurants, sending them when your locations are closed or will be soon may not make sense. This is especially true if the primary call-to-action of your email is to visit a store.
• Call center hours. Similarly, if your emails tend to generate calls to your call center for fulfillment or customer service, for instance, then you probably want to limit your optimal send times to only hours during which your call center is open. You might also be mindful of staffing limitations during various times of the day.
• Flash sales. In cases where there is a small window of eligibility for the sale or event you’re promoting, sent time optimization might not be worthwhile. If you’re running a three-hour flash sale, then your send time window by necessity during the hours right before that sales goes live.
• Limited-quantity releases. Similarly, if you’re selling a limited quantity of a product, then your send time is pretty much prescribed by when that product becomes available for purchase. For instance, if you’re promoting tickets to an event that’s likely to sell out quickly, optimizing your send time will put some of your subscribers at a disadvantage, potentially angering them because they were emailed late in the day after most or all tickets were already sold. Black Friday, when there are a limited quantity of doorbuster deals, is another time when send time optimization doesn’t make much sense.
• Multiple daily sends. Emailing your subscribers multiple times per day can complicate the send time optimization algorithm or testing process. If this describes your email program, consider selecting an optimal send time based upon portions of the day during which you are spreading the sends out. So, for instance, you’d have a best time to send emails during the morning hours and a best time to send emails in the afternoon hours for each subscriber.

Finally, it’s worth stressing that send time optimization is best used with broadcast and segmented campaigns. Automated and transactional emails have their own optimal send time based on the behavior that triggers them. Note that if some of your transactional and other trigger messages are batched and sent at a certain time, you may also want to avoid that time window when sending your other campaigns. 

So, focus your send time optimization efforts on your promotional messages and see what lift they can provide! While timing isn’t everything, when it comes to marketing and driving customer engagement, timing is a major contributing factor to success.

Want more ways to uplevel your email marketing and avoid settling for good enough? Check out:

• When Good Enough Shouldn’t Be: Optimizing Automated Emails
• When Good Enough Shouldn't Be: Fixing Email Performance Problems

—————

Need help taking your email marketing program to the next level? Oracle Marketing Cloud Consulting has more than 500 of the leading marketing minds ready to help you to achieve more with the leading marketing cloud, including strategists, designers, copywriters, trainers, deliverability experts, and more.

Learn more →