Email Testing and Optimization
If you’re not doing email testing and optimizations before you send them to your lists, you’re in the majority of other businesses.
Over 83% of marketers do little or no optimization on their email marketing campaigns. That means that the remaining 17% of marketers are working hard before they click “send” to ensure that their emails arrive in their recipient’s inboxes. As a result, those marketers have higher rates of opens, clicks, engagement and conversions.
With the same kind of work, you can see the same kinds of increases. But you have to do the work first.
Optimization means improving every aspect of your email to ensure delivery, opens, clicks and engagements. If your email isn’t working as well as you’d hoped, it’s time to examine what’s going on, and how to fix the problem.
As a marketer, you’ll focus on three points:
- Understand key behaviors and interactions through email that identifies opportunities
- Identify the “friction points” in your email marketing and your database so you can fix them
- Reviewing your email marketing objectively so you can experiment with solutions to deal with new challenges
Improving your emails before you send them as well as after can help increase your open and click-through rates by as much as 50%.
There are four basic steps to email testing and optimization:
- Segmenting your email list
- Identifying the opportunities in your sales funnel
- Creating experiments
- Prioritizing your experiments
Why would you want to “test”? Perhaps you know how to design, create and write emails, but you may not create something that actually gets opened, read, clicked and converted. Testing allows you to see what works, what doesn’t, and offers you the opportunity to keep trying new things. Even though you’ve gone over and over your emails, if it doesn’t bring in leads, contacts and ultimately, conversions, it’s not working.
Conversion rate optimization (CRO) forces you to concentrate not on the percentage of opens and clicks, but the rates that indicate nothing happened. For instance, you may think that your 25% click-through rate is a good return, but that means 75% aren’t clicking your emails. At this point, it’s time to do some analysis and figure out how to improve your click-through rate.
Remember, too, that most database lists decay at about 23% per year. Many marketers see the solution as generating more leads and growing the list. That’s a good idea, of course, but you should also spend time working on engaging your current leads and customers more. What additional relevant offers can you make to them, and keep them around?
But you also need to think about the users who are leaving you. Constant Contact’s survey asking users why they opted out or marked an email as spam discovered that 69% said “too many emails,” and 56% said “irrelevant content” was the reason. Email optimization can solve these problems for you. Let’s look at each step.
At a minimum, your list should be segmented into customers and leads. But also use other segments that are appropriate for your business. Business personas come into play here, too, as you “divide and conquer” your target market into the ideal customers that you’re selling to. Depending on your business, you might segment by age, gender, nationality, or other company- or industry-specific factors that define your customer.
For instance, if you’re an online pet food and supplies vendor, you’ll want to make sure your “cat people” are separate from your “dog people.” You don’t want to send emails for bird food and supplies to either one of those groups. What would they do with something like that?
If it happens once, and it’s a mistake, that’s one thing. But if you regularly send out bird or fish emails to the cat and dog people, they’re eventually going to stop reading your emails or unsubscribe altogether. Make sure your cat people get relevant emails, as well as the dog people. This is why segmentation is so important.
Emails that are not opened and read indicate unengagement. These addresses should be suppressed going forward.
A/B Email Testing
After dividing your list, you’ll send your current email to one half, and an updated email with one change in it to the second. Of course, before you send, you’ll want to make sure it’s going to the right person, at the right time, with a clear CTA, and make sure the recipient trusts you enough to follow the CTA.
Identify Available Opportunities In Your Sales Funnel
Take time to examine the CRO in your data, find the weak links in your marketing funnel, and why and where that funnel narrows. You’ll use the information you gather to create powerful improvements, prioritize them and create a testing solution.
With email, you’ll use your A/B testing on parts of the email that you can test, such as the number of emails sent, day and/or time of the week, subject line and preview text. Then there are the parts that affect open rates, such as body copy, layout, images, CTA and signature.
This is where your analytics come into play. Examine each of these aspects, and see if there is a pattern that tells you why a click rate is lower than it should be, or why it’s higher. Does one subject line perform better than another? Is the open rate better for mobile or non-mobile readers? Does a particular template or subject wording work better than the rest? Keep track of all these things going forward.
The Holes In The Boat
As your prospects are going through your funnel, you’ll start to notice how many are dropping off on the way to conversion. You know from experience that not every prospect is going to convert. But when your conversions are a miniscule number compared to the number of emails on your list, there are holes in the “boat” that your prospects are riding in. What’s causing them to jump off?
Your goal is to not only identify the “friction points,” or the things that stop someone from proceeding, but identify “fluid points,” or things that keep them moving along the funnel. You not only figure out what you’re doing wrong, but what you’re doing right, so you can keep doing it.
Once you figure out what’s causing these leaks that these leads and prospects are falling through, you’ll work to fix them. You should be able to find at least a few clues, and develop a theory.
- Your emails may be sent at an inconvenient time for the readers to open and read them. (You can read more about that in this article.)
- You’ve sent emails to people who primarily read them from a mobile device, and your message isn’t mobile-friendly.
- Your subject line doesn’t entice the reader enough to open the email to read.
- Your subject line contains emojis, exclamation points or words that set off a recipient’s spam filter (i.e., “FREE!!”) Use a tool like IsNotSpam to check and see if your email will behave like spam once it gets to an inbox.
- Your email ends up in the “Promotions” tab for your Gmail users instead of their Primary tab, where they will more likely see it. Your email should be more personalized, less “business” and “sales-y.” Remove most or all of your images, and have only one link. Use Limtus’s tool for Gmail testing before you send to make sure your email will end up where you want it to. This is especially useful if you have a large number of Gmail addresses.
- You’re sending too many emails. Reduce the number of emails you send, down to one a month or even one a quarter, and see if your open rate improves.
Consider these three things: Potential, Importance and Ease. Which one would have the biggest impact on your marketing funnel? Which would be the easiest? What form of email would you test–your regular e-newsletter, or a single-event email? Consider which one would make a big impact, small impact, or an impact on your email marketing somewhere in between. Once you’ve put together your framework, it’s time to start testing.
Note that testing isn’t something you run once and get all the data you need. A regularly sent email (such as your e-newsletter or an update on an offer) would be better for testing. It should be sent out regularly to get a clear picture of what you need to change or keep. Ideally, a business quarter would be a good length of time to do A/B testing, regularly changing just one element.
Kissmetrics’ A/B Significance Testing tool offers a quick check on how well your testing rated. Keep track of your results until you have enough sampling to determine if your testing worked. Again, you’ll need to run testing more than once or twice to get a good idea of what you need to change, and where. One or two samplings won’t give you enough data to work with.
While optimization is great on its own, email testing your current best-performing templates with different delivery variables or against a different version will help you find a better way to send emails. Don’t just look at your email and think about “what’s wrong with this,” or change an image. Use your data to figure out the best way to proceed.
With a little analysis and configuring, you’ll see what works, what doesn’t, and what works better to increase conversions.