[Study] Email Marketing: HTML or Text?

Email Marketing Advice

It’s one of the most debated questions in the world of modern marketing. When marketing to customers electronically, should we send emails filled with colorful graphics in order to appear more professional and increase our chances of grabbing the customers’ scarce attention? Or, should we send simpler emails containing mostly text so as to appear more personable and less like marketers? To which type of email are customers most likely to respond favorably? Which type of email generates more revenue? That’s the question.

In a fascinating conversation with email marketing expert DJ Waldo, online radio host Todd Schnick asks just this question. DJ’s answer, not surprisingly, is that it depends on the audience. He offers Chris Brogan and Derek Halpern as examples of people who do text-based emails really well and the clothing company Ibex as an example of a company that does image-based emails really well.

In his book The Rebel’s Guide to Email Marketing, coauthored with Jason Falls, DJ explains that the default answer to the question for most marketers is typically, “HTML.” If we have the capability of making our emails look fancy, why wouldn’t we do it? DJ offers some possible benefits of using text-based email rather than image-based emails:

  • HTML emails may display differently on difference devices, while plain text emails will be consistent across all devices.
  • HTML emails contain security risks, as they are often used for phishing scams, and may be considered SPAM by some email service providers.
  • Plain text emails are more personal in nature, causing recipients to associate the sender as a friend or colleague rather than an advertiser.

In the end, though, DJ admits, “The key to success is to test. What works well for one audience might be totally different for another.”

The following study seeks to take this question in another direction. Like DJ Waldow’s insights, most research up to this point has focused on the type of audience as the key criteria for deciding between images and text. This study seeks to put the question in a new light by comparing email marketing campaigns, not against different audiences, but rather against the same audiences in different stages of the buying cycle.

Article Summary: Customer Relationship Stage and the Use of Picture-Dominant Versus Text-Dominant Advertising

Lewis, M., Whitler, K., & Hoegg, J. Customer Relationship Stage and the Use of Picture-Dominant Versus Text-Dominant Advertising: A Field Study. Journal Of Retailing. September 2013;89(3):263-280.

Michael Lewis of Emory University, Karen Whitler of Indiana University, and JoAndrea Hoegg of the University of British Columbia collaborate to investigage a number of theories regarding the impact of customer relationship stage on the effectiveness of email marketing. Based on both prior research and simple intuition, they come up with the following factors that may influence the decision:

  • Whether the recipient is a prospective customer (has not made a purchase) or a returning customer (has already made a purchase).
  • The length of time the recipient has been receiving emails (amount of prior exposure to messaging).
  • The level of involvement the recipient has with the message contained within the email.

The Experiment

To test their hypotheses, the researchers examined the email marketing campaigns a large women’s fashion retailer. Using the same copywriter and graphic designer for both campaigns, and crafting the same message, the researchers constructed a picture-dominant email and a text-dominant email for two different campaigns:

  1. Prospective Customer: this campaign was sent to people who had not yet made a purchase, promoting an item commonly bought as an initial purchase.
  2. Experienced Customer: this campaign was sent to people who had already made the initial purchase, promoting an item that is commonly purchased following the purchase of the item described above.

Picture-Dominant vs Text-Dominant Email Template

For the prospective customer campaign:

  • The text-dominant email contained 191 words more than the picture-dominate email.
  • 21% of the space was devoted to pictures in the text-dominant email versus 61% in the picture-dominant email.

For the experienced customer campaign:

  • The text-dominant email contained 201 words more than the picture-dominate email.
  • 60% more of the space was devoted to pictures in the picture-dominant email than in the text-dominant email.

After pretesting the emails to ensure the distribution of text and images was consistent with other national campaigns (and confirming prior research suggesting that image-based advertisements generate a more pleasant emotional response than text-based advertisements), the researchers sent out the “Prospective Customer” email to 34,563 recipients and the “Experienced Customer” email to 17,984 recipients.

The Results

After the test campaigns were sent out, the following data on email response was gathered.

Email Response Rates

As far as email response goes, there is not much difference between text-dominant and picture-dominant emails for either prospective or experienced customers. The one significant difference that the researchers highlight is that, for prospective customers, picture-dominant emails have significantly more clicks per open than do text-dominant emails. This result implies that, for prospective customers, image-heavy emails are more likely to increase their depth of interest than text-heavy emails.

The email response behavior, however, is relatively inconsequential. The more significant results have to do with the purchase behavior of the recipients. Initially, it seems that there is no significant difference between purchase behavior in either “prospective customers” or “experienced customers” when exposed to either text-dominant emails or picture-dominant emails. However, when the researchers further segment the lists based on recipients’ propensity to open emails, they find that the choice between text-heavy and image-heavy emails is critical when it comes to “prospective customers.”

  • For new customers with high email open rates, using text-dominant emails increases revenues by 8.3% as compared to using picture-dominant emails
  • For new customers with low email open rates, using picture-dominate emails increases revenues by 5.9% as compared to using text-dominant emails

Real-World Implications

If you’re still with me, you’re probably thinking that I’ve built up a bunch of hype about this study with little to show for it. In a sense, you’re right. The study does not reveal dramatic differences, as far as generating revenue, between using text-based and image-based emails in email marketing. But, I nevertheless think that it’s a pretty big deal to have this fact validated scientifically. It turns out that DJ Waldow is right; you’ve just got to test campaigns and see what works. There are no rules.

That being said the study does highlight a pretty fascinating distinction between the purchase behavior of those with low email open rates and high email open rates in regards to the text-to-image ratio in the emails they receive. Also, a significant point was made in regards to the effect of image-heavy emails on clicks per email open. So, at the end of the day, here are some takeaways for anyone using email marketing to reach customers, donors, subscribers, and so on…

3 Takeaways on Email Marketing for Potential Customers

  1. To generate more engagement (clicks per email) from potential customers, use image-heavy emails in your campaigns.
  2. To generate more revenue from potential customers with high email open rates, use text-heavy emails in your campaigns.
  3. To generate more revenue from potential customers with low email open rates, use image-heavy emails in your campaigns.

Questions for Future Research

  1. Would the results be different for a different industry? Perhaps for an industry with less visual appeal than apparel?
  2. Could there be other online variables significantly effecting the results? Like social media commentary, blog posts, etc.?
  3. Is there something unique about the electronic medium? Would a direct mail marketing campaign with the same image-to-text components yield the same results?
  4. In the “clicks per email” metric, are the recipients clicking hypertext or images? In other words, which is more clickable, images or text?

featured image courtesy of Mike Schinkel licensed via Creative Commons


About Douglas E Rice

Douglas E Rice is just a guy who likes to learn stuff.
This entry was posted in blog, Marketing, Studies. Bookmark the permalink.

One Response to [Study] Email Marketing: HTML or Text?

  1. geoffreycallaghan69 says:

    I don’t think email marketing is as effective theses day
    I would try a http://www.TeenyText.com service like TeenyText.com

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s