How to Analyze & Optimize Your Email Automation
Editor’s Note: This post is the last of a three-part series explaining how to set your business up for success with email automation. Be sure to also read how to develop a data-driven automation strategy & then how to bring your strategy to life with targeted customer journeys and relevant content.
You’ve done your homework and come up with a solid email automation strategy. You’ve also implemented that automation strategy by mapping out and building each autoresponder, populating them with relevant content, and turning them on. Now you’re ready to sit back, relax, and watch your business reap the benefits, right?
If only it were that easy! Embracing a “set it and forget it” mentality off the bat can lead to big misses, like overlooking a flow that never actually started sending. Over the long term, if you fail to check on your flows periodically, there are also major downsides. You risk sending outdated content that has lost relevance due to changes with your business or products, and most importantly you miss out on opportunities to test and optimize for better performance.
In this post, we cover how to monitor the performance of your email flows over the short and long term. We’ll also share a list of easy experiments you can start running today to optimize any automated sequence.
Monitoring performance not only allows you to identify areas for improvement in the short-term, it also allows you to gain deeper insight into your business on an ongoing basis.
For each automated flow you implement, you’ll want to look performance across four key indicators:
Delivery Rate. How many people are qualifying for each flow on a daily or weekly basis? If these numbers are lower than you expected, consider whether your qualifying trigger is too exclusive. Realize that for some behavior-triggered sequences, low volume could also be a good thing! If an Abandoned Cart sequence isn’t very active, this could mean you don’t have many people starting the checkout process at all (not so great…) or simply that most people do complete their purchase and thus won’t qualify for an abandoned cart reminder (much better!).
Open Rate. Are people opening your emails? If your open rates are below 20%, try A/B testing your subject lines to stand out better in a crowded inbox. Also, consider auditing how many emails your recipients might be getting from you each week; using a tool like Klaviyo’s Smart Sending can help you avoid sending too many emails too often. If open rates are consistently low, it’s also possible your sender reputation is landing you in the spam folder and you need to examine your email marketing practices.
Click Rate. If your open rates are high but your click rates are low, this means you have some work to do on the content front. Are your “Calls-to-Action” (CTAs) clear in each and every email? Are your communications relevant and personalized? Automated emails in particular should be highly relevant, as they are triggered by someone’s behavior. Consider ways you can get more targeted with each communication to maximize engagement.
Conversion Rate. Not all emails are designed to get someone to buy, so consider what conversions you want to measure for each email or email sequence. If you’re hoping to drive blog traffic, for example, measure how many people click through and stay on your blog for more than 10 seconds. If you are focused on revenue conversions, look at how many people open vs. click vs. purchase after receiving the email. High opens and clicks, but low order conversions, means you may need to focus on optimizing your website and/or checkout experience to reduce friction.
You should also track how much of your business revenue is coming from certain autoresponders, and from your marketing automation as a whole. If these numbers start to dip, you should go back through all the key stats above to figure out where in the funnel people are dropping off. Are you seeing less people visit your website overall, and need to focus on top of the funnel? Are you seeing an influx in one-time-only buyers, and need to invest more heavily in retaining existing customers and encouraging repeat purchases?
Keep in mind that depending on the size of your business, you may need to wait several weeks until your new autoresponders have seen enough volume to reach statistical significance. By checking performance on a weekly basis, you can watch trends over time and wait to fully analyze results until your key stats start to level out and remain consistent.
Tweak, Rinse, Repeat
When you start monitoring the performance of your autoresponders, invariably there will be areas for improvement. Some may be glaring, like a cross-sell sequence leading to a 0% conversion rate because you forgot to add a clear CTA in every email. Others may present themselves more as areas for optimization or experimentation.
There is often little to lose, and a lot to gain, by initiating controlled experiments to see how you can optimize an individual email or even an entire series.
Here are some ways you can start experimenting with different optimizations:
- A/B Test Subject Lines. Experiment with tone in your subject lines. Does your audience respond well to emojis? You’ll never know unless you give it a try! A/B test different subject lines to see what draws people in. Suffering from writer’s block? Checkout these 10 techniques to start writing better subject lines.
- A/B Test Timing. For sequences designed to send a specific timely reminder (perhaps to revisit an abandoned cart, or come back to buy a second time), you’ll likely want to run an experiment to figure out what “timely” truly means. Split out 50% of recipients and send them down a path with Timing A, where the other 50% get Timing B. Continue testing until you feel confident you’ve hit a sweet spot with the right wait periods.
- Experiment with Plain Text Emails. Consider trading some flashy HTML email templates for simple, plain text messages that have a more personal touch. For certain buyer personas or target lifecycle stages, opting for a more personal sounding message from a key person at the business — presented as if it were written from that person’s own inbox — can be really effective at driving engagement. These types of emails also have a much better chance of landing in someone’s Primary inbox, while image-heavy emails will almost always land right in Promotions. At Klaviyo, we’ve seen this technique work really well in a Welcome Series or an Abandoned Cart reminder.
- Experiment with Different Incentive Structures. A/B test adding richer discounts or incentives into certain sequences where you want to drive up revenue conversions. In advance, calculate the conversion rate you’d want to hit to make the incentive offering worthwhile in terms of ROI. Start with a small test group (think < 20%) if your confidence level is low.
If you do find yourself happy with your results, consider extending or expanding your marketing throughout the customer journey, integrating new flows or more robust content. Conduct another audit to identify gaps or weaknesses. You can also start to strategize around the other ways your subscribers engage with your brand or business, like through social media or cross-channel advertising. Consider auditing these other experiences to ensure you are providing a seamless customer journey across all channels and touch points.
Well, what are you waiting for?
Getting started with email automation can seem like a daunting task. By following the best practices outlined in this three-post series, you’ll be able to get your feet wet while knowing you’re investing your time wisely.
Most importantly, don’t forget to keep the customer journey as your guiding light as you work to automate more and more key lifecycle stages. If you do, and keep an eye towards monitoring and optimization, you’ll see your strategy pay dividends to increase the lifetime value of your subscribers and customers.
What has helped you make the most of marketing automation? Let me know in the comments!Back to Blog Home