Improving on Email Marketing

  •   Doug Logan

By Doug Logan

There was once a day that email marketing was worth its weight in gold, but those days are long gone. So, what do we do now?

So, what do we do now?

Marketers moved from sending tons of printed mailers to sending more and more highly targeted and personalized emails. But the day when consumers read every email received is long gone (which is one of the many reasons why print is not dead, but that's a story for another blog post).


The Format.

Think about the emails you receive every day. They mainly come in one of two formats. You have emails like newsletters and promotions that have a level of design and images (maybe even some subtle animations) which I will refer to as "graphic emails".

The other kind of emails are the ones that you get from your colleagues, friends and family. Work-related emails from coworkers, the latest forwarded joke from your friend and that awesome email from your Mom asking about those awkward pictures on Facebook from the weekend (sorry, Mom). The emails typically lack design and are heavily text-based, so I will refer to them as "text emails".


The Test.

We had just finished up launching a new website for one of our clients and wrote up the case study for the project, as well. Normally when we have something new to share, we broadcast it in an e-newsletter (a graphic email) to our list of subscribers as well as via all of our social networks. But this time we thought, why not test which of the two formats (graphic or text) is more effective at driving readers to click through and read the case study?

We took our master list of ledes, clients and general contacts and split it into three groups. Each group is roughly the same size with the smallest list being less than a hundred subscribers smaller than the largest list. All the emails went out on the same day (a Thursday).


Test Group A.

This group consisted of half of the most recent and active contacts in our database. We sent them a graphic email that looked like this:

Graphic Email

Test Group B.

This group was comprised of the other half of the most recent and active contacts in our database. They were sent a text email that looked like this:

Text Email

Test Group C.

This group was comprised of older contacts that signed up for our e-newsletter and we have had very little interaction with. They have been in our database for years and were sent the exact same graphic email that Test Group A was sent.


The Results.

Before we conducted this test we had a gut feeling that Test Group B would outperform the others, due to the fact that the text email was more personal and shorter, but when we got the results we were quite amazed.

Test Group C only had a 2.8% clickthrough rate (still higher than the industry average of 2.4% according to MailChimp). Test Group A had a 6.5% clickthrough rate and Test Group B had an astonishing 14.3% clickthrough rate.

That means that the text format was more than twice as effective at driving readers to the case study then the graphic format.

Now of course there are more variables that factor into this, such as the content changing due to the format. I would have also liked to do a double test where we sent Test Group A the text version and Test Group B the graphic version, but for as rough as this test was, I am still very happy with the results. Hopefully this will encourage you to start testing your email marketing efforts and, if you try this same test, please feel free to share your results!

Header photo by Gajman

  • Dave Morse

    Interesting test, for sure. But it seems inconclusive to rest on the notion that Group B outperformed simply because it was a "text email" (personal, shorter) vs. "graphic email". I'm general, I'm a big proponent of sending information that I think will ADD VALUE to the recipient. So, if it achieves this goal, then it would likely perform well in either text or graphical format. My .02


    • Doug Logan

      Very good point Dave. It truly is an inconclusive test at this point and does need some further testing to validate this theory. However, I will point out that information was the same in both versions of the email just presented differently. Thanks for your thoughts!