Email Marketing: What to Test and How to Test
One of the most powerful aspects of email marketing is that it can be tested (and thereby optimized). Following generic “best practices” or fixed “rules” is not a path to success, because what works for one audience won’t necessarily work for yours. Testing is your biggest ally. You should always ‘test for the best’ in your email campaigns.
What to Test
Historically, the subject line has been the most popular marketing email element to test, followed closely by the content of the message itself. But as MarketingSherpa highlights in this chart, there are many other elements of your emails that are just as testable.
Here are 25 elements you can consider testing:
- Subject line
- From name
- Day of the week
- Time of day
- Mostly-images vs. mostly-text
- Short copy vs. long copy
- Links vs. buttons
- Number of links
- Unsubscribe at the top
- First name personalization — in the subject line
- First name personalization — in the email body
- Animated gifs
- Font colors
- Font styles
- Social sharing icons
- Social connecting icons
- Delivery by time zone
- Call to action — number
- Call to action — placement
- Post-click landing page
- Social proof
- Tone — human vs. corporate
- Copy length
Here are some results of real email tests we’ve run at Marketo.
How to Test: Eight Pro Tips
- Start simple. Test subject lines and headers first. It doesn’t take a lot of time or creative work to come up with a few simple variants, and the return can be significant.
- Test one element at a time. If you test more than one element, you won’t be able to tell which variant drove the success.
- Control for time of day and day of the week. If you’re testing other variants, then send on the same day and at the same time to eliminate the timing variant.
- Keep a log of all your tests. Record your findings so you can refer back to the specific variables tested and, more importantly, learn from them.
- Make sure testing is part of your day-to-day processes. Testing doesn’t have to be daunting, and it shouldn’t be something you put off due to a lack of resources — it should be part of your daily routine.
- Run tests on groups that are small, but large enough to determine a clear winner. The winning variables should then be incorporated into your larger mailing.
- Don’t forget that small differences can be significant. This is especially true if your sample sizes are large. Visit Marketo’s The Ultimate Guide To Test Statistics to learn all about this. And for those of you who self-proclaim, “I’m not good at math”, we’ve got a handy A/B testing calculator you can use. (This calculator was originally designed for landing-page conversions, but it works great for email numbers. Just replace “page views” with “emails sent,” and “conversion rate” with “click-through rate” or “open rate”.)
- Listen to what your tests tell you! All the testing in the world won’t matter if you’re not making decisions and modifications to your campaigns based on what you’ve learned. Unused data is sad data, indeed.
Enter the WhichTestWon Email Testing Awards
Speaking of testing, Marketo is a proud sponsor of the 2013 WhichTestWon Email Testing Awards. Entries are due by 5 pm PST on Tuesday, July 30. Categories include:
- List gathering: opt-in pages, forms, offers or campaigns
- Landing pages: specifically for email-driven clicks
- Newsletter template
- “Blast” template – offer campaign
- Autoresponder or triggered campaigns
- Transactional notices
- Subject line or “from” name test
- Mobile test
- Social test
Go ahead and enter your test to win fame, links, and a shiny blue ribbon!