5 Easy Email Marketing A/B Tests
Testing is one of the secrets to ecommerce success. But if you’re on a deadline or have limited resources, the pressure to get the email message done may outweigh all other concerns.
The good news is testing doesn’t have to be hard. There are plenty of simple A/B tests that deliver good results with very little effort. You won’t need an advanced degree in testing methodology or fancy analytics skills. And the best part is the results. These tests have made a major difference for thousands of email marketers — they will move the needle for you, too. Here are five easy tests you can run to improve your email marketing results.
1. Subject Lines
Subject line tests are the simplest, fastest tests. They can generate good results.
You get two rounds of benefits from subject line testing. First, you’ll see results within 72 hours of when you send the two versions of the email — that’s when the bulk of the responses will come in. Later you can run a report to see subject line performance over the last month. Look beyond the most basic metrics like open rates and click rates to actual revenue. Find out which email subject line generated the most revenue.
2. Call to Action
The call to action is where the rubber meets the road, so to speak. Get your call to action right, and you’ll have a successful email on your hands. Get it wrong, and your email marketing sales will fizzle. The whole point of marketing is to get your subscribers to act, so testing the call to action is always a good idea.
Make your call to action clickable and enticing no matter what. If a subscriber has images turned off — and about half of them do — they can’t click a call to action that’s an image. So play it safe and have all the buttons in your emails coded in CSS — not created as images. With CSS your subscribers will see a nicely-formatted, eye-catching button no matter what their email client image settings are set to.
Have you, like so many other marketers, collected more than an email address on your opt-in forms, but rarely, if ever, use the data you’ve collected?
It’s time to start using all the personalization data you’ve been collecting. Try adding a dynamic field, even if it’s just adding the subscriber’s name to the salutation. MarketingSherpa reported a 17.36 percent higher average click rate across 7 personalized subject line tests for a B-to-B email marketing campaign. ExactTarget, an email service provider, posted on its blog, “Adding personalized recommendations into marketing emails can increase sales conversion rates by 15 to 25 percent, and click-through rates by 25 to 35 percent.”
With results like that, it’s got to be worth a try.
If you’re scared of botching the dynamic field, try a test mailing to just a sample of your list. A sample of 10 to 20 percent of your list is a safe place to start. That way, if you do make a mistake with implementing personalization, at least you’ve minimized your exposure.
Sending an experimental email message to a small segment of your list is a good way to hedge your bets. One thing testing advocates rarely talk about is what happens when a business owner loses money by doing a test — for example, when the test cell doesn’t beat the control, and the owner loses a chunk of sales as a result.
If you simply can’t lose sales, even for the sake of long-term results, mailing to a small segment — again, around 10 to 20 percent, not 50 percent — will minimize the damage if your risky test goes wrong. And all tests have some risk.
4. Delivery Times and Days
Timing may not be everything, but it sure can help — as in “double your click rates” kind of help.
There are dozens of reports on which times and days are best to send emails. Unfortunately, the reports give conflicting information. The only way to know what will work best for your list is to test it yourself. (I’ve previously addressed email delivery times, at “When Is the Best Time to Send an Email?”)
This brings up another little secret about testing. You have to retest key elements like delivery times and ongoing offers every 6 months or so. So if you haven’t tested which delivery days or times get you the best results in the last 6 months, you’re due for another timing test.
5. Your Mobile Formatting
About half of your emails will be opened on a mobile device. Even if you are up to speed on how to make your emails look good and perform well on mobile devices, odds are some additional testing could help. Everything you do to squeeze more juice out of mobile puts you that much further ahead of your competitors.
Here are two ideas for what to test in your mobile format.
- Load times. Will the faster load time of an email with a simpler format and scaled-down images help your sales? Strangeloop, a web optimization firm, reports on its blog that “74 percent of mobile users say they will bounce after waiting more than 5 seconds for a mobile page to load.” Yet “the average m-commerce site takes 9.86 seconds to load.”
- Text size. Will an email with larger fonts — like 24 for headlines and 14 for body copy — outperform an email with smaller fonts?
Running even one of these tests a month can make a dramatic difference to your email marketing. If it still seems like too much to do, ask yourself what it would mean to your business if you got 30 percent more results out of your email program. What if you saw 50 percent more results?