How often should I email my subscribers? Without a doubt, that’s one of the top five most frequently asked questions my clients have asked me over the years. My quick answer?
If the content is relevant, then frequency is irrelevant.
Put another way, if your emails contain information, offers, news, updates, and other content your subscriber wants, needs, or expects, then you can essentially send as often as you’d like. That’s the ideal, but the concept of an always on, completely fluid construct of emails that are triggered as necessary where the right content is always available and automatically pulled when needed simply isn’t possible today. AI and other technologies are helping marketers get closer to that ideal, but the reality is that it’s many years off for most email teams right now.
So, while we wait for the future to arrive, what’s the next best thing? How do I know what the best email frequency is?
First, it’s important to start with the customer. This rocks some worlds but...the customer may not want as many emails as you are sending. That can be tough for many brands to accept because cutting back email volume is associated with cutting back revenue. Some see the “send button” as the “revenue button.” While it’s true that every time you push send, conversions and revenue spike, there are many long-term negative implications of over-mailing, including high list churn, poor deliverability, and brand damage.
But you can avoid those negative consequences while not forsaking your short-term revenue goals by being thoughtful about figuring out which subscribers should get each of your non-triggered promotional emails. The frequency of promotional emails across the industry continues to grow, so the risks are growing, too. Ironically, volumes are growing in part because of the ongoing effectiveness of email.
So what are the best ways to smartly, even surgically, select who should get those emails that are sometimes sent multiple times in a day? In keeping with my previous posts on the best time to send emails, optimizing automated emails, and fixing email performance problems, let’s look at the Good, Better, and Best approaches to email frequency:
Most businesses have some cyclicality to them, and those seasonal pushes often mean boosting email volume by 3x or more during peak periods. This spike may be a function of any or all of the following tactics:
Incremental sends, which involve sending additional emails, often multiple times per day
Re-sends, where the same email is sent multiple times, often only with a change to the subject line
Deeper audience selection, where emails are sent to a broader, less active audience of subscribers that include subscribers that you typically suppress from sends due to their lack of activity
Each of these is a common approach to increasing volume in an attempt to maximize revenue during peak seasons. However, applying the right rules is important to reduce the risk of jeopardizing your subscriber base long term or your IP reputation and associated deliverability success in the short term. Let’s look at each approach separately:
For B2C brands, the November to December holiday season is the most common period where they increase their email frequency above their typical levels. Sending two or even three times a day isn’t uncommon for many retailers. Whenever your peak seasons are, you’re likely increasing your email frequency significantly to try to capture your customers’ heightened interest.
Reduce the risks of using this tactic by being selective about which subscribers you target with incremental sends. Ask yourself…
Which segments of my audience are the most likely to engage with me during a particular peak season window? Examine previous periods. How many subscribers didn’t open any emails at all during that period? Could that have been predicted to begin with?
Which subscribers have a pattern of only engaging during that time of the year?
Can I conduct a matchback to offline activity via email address, loyalty program, or some other approach to understand the impact of offline conversions? As we know there can be a halo effect associated with email impacting store traffic even without opening or clicking on an email.
Answering those three questions will help you focus your incremental sends on subscribers who are likely to act, either online or offline, and avoid sending incremental emails to those who are unlikely to act.
If you've not tested the impact of this across different segments, try it out. Re-sends can go to recipients of an email that:
Didn’t open or click
Opened but didn’t click or convert
Clicked but didn’t convert
Based on experience with our clients, non-openers are the least likely to respond to re-sends, especially if they haven’t engaged lately. For example, if your brand sends at a high frequency and a subscriber hasn’t opened any of your promotional emails over the past 30 days, the odds of them converting on a re-send of an email on the same day are extremely low.
Reduce the risks of using this tactic by thinking of your re-sends like browse abandons. For those subscribers who demonstrated that they were reasonably serious by clicking but not converting, target them with a re-send that highlights a time-sensitive aspect of the promotion, such as “Just hours left until your 50% off promotion expires!”
It’s pretty simple to test and see for yourself how each of these groups responds to re-sends and where the biggest conversion opportunity is.
Mailing subscribers who haven’t heard from you for some time can be trickier to navigate for several reasons. You haven’t been mailing them recently because their engagement level was so low that it was hurting your engagement and you didn’t want it to affect your deliverability. So, it comes to reason that starting to mail them again presents similar risks. You can expect:
Spam complaints to rise
Engagement rates to dip
Potential bulking or blocking by inbox providers, because of those first two issues
Reduce the risks of using this tactic by being selective about going broader with your audience. For instance, you can reduce your deliverability risks by only adding inactive subscribers who have made purchases in the past.
You can also reduce the fallout from loosening your audience criteria by reaching out to these subscribers well ahead of the holiday period. Try to warm up this expanded audience by including them in some campaigns leading up to the core season, so if deliverability issues do appear, you have time to mitigate them rather than trying to solve that during the mission critical period.
Established brands generally have 18 months or more of email engagement history for the majority of their subscribers. This is a gold mine of data that can be used to generate basic segments that indicate value metrics associated with recency, frequency, and monetary (RFM) dimensions of engaging with you.
RFM segments, as they are known, have been around for many years, but broad application into email is more recent. Some digital marketing platforms and email service providers, including Oracle Responsys, have RFM values built in. But if yours doesn’t consider creating them manually for your file. Subscribers are behaving very differently, so we should treat them differently—and this is a great rubric for doing so.
Consider the differences in the following audiences:
Hyperactives. Based on the last 6 months of send data, these subscribers have opened or clicked on 80% or more of emails you’ve sent to them. This data is even more helpful if you can layer on a value dimension—the monetary one—to understand which of your hyperactives are generating the most value for your brand. This is an audience you may not be mailing enough! The focus here should be on further personalization and loyalty expansion.
Zombies. Based on the last six months of send data, these subscribers haven't opened or clicked on anything. However, they were high-value subscribers at one point in time. Is it possible to bring them back from the dead? These are prime re-engagement targets that warrant a different frequency and content strategy.
First-timers. Based on six months of data, these subscribers were moderately engaged in terms of recency and frequency of engagement early on, but haven’t opened, clicked, or converted since. This audience is a prime target to turn into first-time purchasers or repeat buyers.
Several other behavioral segments can be created to join these three, depending on how your audience behaves and what the natural buying cycle is for the products or services your company sells. Note that not only is the frequency likely different for these audiences, but the content and messaging will become sharper and more relevant based on what we know about them.
Finally, given that we’ve done our homework with these audiences, the more that the messages sent to them can be trigger-based, rather than simply a part of the natural promotional calendar, the more they are likely to engage with the brand on the optimal path. Think about the push to get a one-time buyer to be a repeat buyer and what you know about the first purchase that could be instructive regarding their second. Granted, lots of factors come into play here—including that you might be selling a durable good that your customer only buys once every three years—but think about what that messaging should be after the first purchase to ensure you are a part of the consideration set for the next purchase.
You knew it was coming: We’re going to talk about artificial intelligence now. All that data we talked about for the Better approach? It’s useful here too, but we can also combine it with other subscriber data like past purchase history, gender, acquisition source, and so on to establish propensities and anticipated behavior to inform frequency.
Most subscribers will go through natural ebbs and flows with a brand. Imagine going on a large vacation at a resort, or purchasing a car. For most people, these are not frequent purchases. From an email standpoint there will be peaks and valleys of engagement based on the natural lifecycle.
A predictive model powered by AI can determine the probability that a subscriber will engage with an email or, even better, make a purchase. Imagine if each of your subscribers was rated from 0 to 100 based on their likelihood to that they would engage with your next email in the next 3 days. If you were considering ramping up frequency from once per day to twice per day, with this rating you’d have the ability to, for instance:
Select only those with a 35% or greater chance of engaging with a re-send
Select those with an 80% or greater chance of engaging with a 3rd send from among those who didn’t engage with the re-send
Know who to target if sales are short at the very end of the quarter to minimize the risk of fatiguing or burning out your email list
We’ve run this kind of predictive activity modeling successfully over and over for clients where it has allowed them to send less email and generate more revenue when compared to a control group that didn’t use the modeling. For example, we conducted a two-month experiment with an outdoor apparel retailer where one audience segment received their regular email cadence and another segment included some recipients received slightly more email and others slightly less. The segment that used predictive activity modeling sent 19% less email but generated:
17% higher open rates
29% higher click-through rates
29% higher conversion rates
Revenue equal to the control group
By avoiding unproductive email sends, brands create better subscriber experiences and are rewarded with fewer unsubscribes and less email fatigue, and therefore higher subscriber lifetime value. Plus, the higher engagement rates improve deliverability, which is especially critical for high-volume senders who receive the most scrutiny by inbox providers.
A lot of factors go into answering How often should I email my subscribers? Seasonal optimization, engagement segments, and AI-powered predictive modeling can all play a role. I hope that the issues I discussed in this post have armed you with new approaches and insights into how to best approach this tricky balance as you head toward your next big promotional period.
Want more ways to uplevel your email marketing and avoid settling for good enough? Check out:
Need help taking your email marketing program to the next level? Oracle CX Marketing 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.
To learn more about email frequency and the tools to make email marketing successful, visit Oracle CX Marketing.
Clint Kaiser is the Head of the Analytic & Strategic Services team at Oracle Marketing Consulting. His background in the email marketing space includes 20 years of experience with ESPs and digital agencies. His analytical approach to driving change in digital marketing is reflected in his quantitative approach to improving clients' business outcomes.