Email testing seems to be a science in itself, and marketers often struggle with identifying strategies to assess deliverability and open rates. The point of testing emails is to experiment with different elements to ensure that campaigns yield the best results and derive the most value.
It is imperative that the tests are conducted and analyzed properly, “There is a cost for bad testing,” says Corey Trent, Research Analyst, MarketingExperiments. “Bad assumptions based on bad tests can cost you a lot of money and cause you to lose out on a lot of business.”
We encourage our clients to implement an email marketing test plan based on best practices and have compiled a list of methods and tactics needed to do so.
Email Testing Methods
There are two main methods for email testing:
1. A/B split tests
2. Multivariate tests
In A/B email testing all elements remain the same with the exception of the one being tested (i.e., subject line, image, offer) and the test should be administered by experimenting with only one variable with each email send. Multivariate email testing is a bit more complicated and involves testing multiple elements in a single send. It also requires a larger list to garner a statistically significant result, and the outcome identifies the specific combination of elements that produced the best response.
Email Testing Elements
Email marketing is comprised of many elements that are ripe for testing including the message, appearance, and delivery. Some things you can test in the message include:
- Subject line
- From name
- Headline/sub-headline
- Content
- Promotion/offer
- Post click experience (landing page A versus B)
When it comes to appearance, consider testing graphics versus text, colors/styles/fonts, and the layout. Delivery elements also afford opportunity for testing as marketers consider time of day, day of week, and email tools.
Email Testing Measurements
There are many ways to measure the impact of email campaigns, but the most vital measurement is the amount of revenue you can generate from one version versus another. In terms of measuring engagement or interest, your best bet is to track clickthroughs, or more specifically, the links that are clicked. It is important to note that opens (versus clicks) can complicate measurement since some recipients may have a preview pane activated that will automatically register emails as opens. And finally, track and measure the number of unsubscribes and monitor the results.
Best Practices for Email Testing
We recently uncovered a fantastic resource for best practices around email testing, The Ultimate Guide to Email Marketing by Kevin Gao, CEO & Founder of Comm100. The main points covered in the comprehensive e-book have been rolled up as follows:
List Segmentation Testing
- Segment list by email service provider
- Segment list by customer behavior
- Segment list by demographic
- Manually check email list segmentation to ensure that you don’t over-email individuals
- Make sure you don’t segment too small
A/B Testing
- Remove inactive users
- Remove highly active users
- Sort alphabetically
- Do not sort by join date
- Split highly active or highly inactive users separately
- Check percentage of email service provider addresses
Copy and Text Testing
- Make sure that headline font, size, and placement is the same for both tests
- Use the same copy writer for both versions (small differences in tone or style can account for variations)
Image Testing
- Don’t make any other variable changes other than image
- Use similar alt and title text behind the images
- Plan out your image test to track what and when testing
From Address Testing
- Make sure your A/B test is clean (since success metric is open rate, your lists must be as similar as possible)
- Test the “from” address early because it may have significant impact
- Test to ensure that your message is being delivered (beware of spam issues)
- Note that “from” addresses may be situational (one may work better for transactional versus marketing emails)
Subject Line Testing
- Make sure your A/B test is clean
- Test concepts over several sends
- Be aware of inbox deliverability
Send Date and Time Testing
- Make sure A/B test is split evenly across the database
- Test several times to ensure unexpected factors didn’t cloud findings (holiday, internet downtime, etc.)
- Think about target persona and when they are online
Offer and/or Promotion Testing
- Consider the value: a less well-responded to offer that makes a profit may be better than a popular office that actually loses money for you
- Test different offers from different portions of your database
- Evaluate whether timing plays a role and run the test several times
By implementing an email testing strategy, you can gain valuable insight in terms of how your campaign will perform. Testing leads to the opportunity to optimize which generally results in better results, so we encourage clients to consider this approach to maximize email marketing campaigns.
Need help with email testing or with your marketing strategy? Contact Launch Marketing today! Be sure to follow us on Twitter or like us on Facebook to keep up with the latest B2B marketing tips and trends.
Graphic Source: https://zapier.com/learn/ultimate-guide-to-email-marketing-apps/ab-testing-email-marketing/
There are no comments