Add Testing To Your Email Marketing Strategy

You probably know you should be doing some level of testing as a part of your Email Marketing Strategy. Running a test and learning from the results will enable you to see what is performing best and allow you to make informed decisions on your future email marketing campaigns.

I have done lots of testing (that is not a boast) which also means I have done lots of the wrong testing. Read on for 8 top tips.


1.   A/B Subject line (and anything)
The subject line A/B split test is the classic split test. It is testing two different subject lines, with the same email creative, and seeing which subject line generates the best response. The subject line is a vital piece of the puzzle to get people to pay attention to your email and stand out in the inbox and also to get your subscribers to open emails (an open being when images are downloaded).

What can you A/B test?

Your subject line should answer the following questions:

  • Who’s it from?
  • What’s in it for me?
  • How do I take immediate action?

Try testing the following:

  • Long vs short subject lines
  • Personalization vs no personalization
  • Brand name vs no brand name
  • Discount vs no discount
  • Capitalization vs lowercase
  • Time limit vs no time limit

I recommend running a subject line test at least once a month. People can say you should be testing every time you send an email but I just don’t think that is realistic.

Remember to not just look at the open rate to determine which subject line performed the best. Also look at click-throughs, web visits and conversions. I have run tests where the open rate was inconclusive but one subject line had significantly more conversions.

The easiest and the standard way to run the A/B test is to split your recipients 50/50. However another option is the Pre-mailing test – see number 4.

2.   Pre-header testing
The pre-header is an extension of your subject line and something that is visible in the major email clients. If used in the right way it can get you more opens and engagement from your email campaigns. Therefore it should be tested!

My guidelines for the pre-header follow the same lines as the subject line and should therefore answer the following:

  • Who’s it from?
  • What’s in it for me?
  • How do I take immediate action?

In addition, I would test the following:

  • Mobile version call out vs main message (see example below)
  • Personalization vs no personalization
  • Brand name vs no brand name
  • Discount vs no discount
  • Time limit vs no time limit

3.   From Name
Ever tested the From Name? Probably not, or less so than anything else. However the From Name is one of the first things your recipient sees in their inbox. Especially in desktop email clients such as Outlook an example of which is shown below:

From-Names-Outlook

So what can you test?

  • Brand name vs Brand email type
  • Brand name vs CEO or contacts name
  • Brand vs Brand and team name

4.   Pre-mailing test on subject line (and anything)
A pre-mailing test involves sending two different subject lines with the same email creative to 20% of your audience (10% receive one subject line. 10% receive the other). Wait 2-4 hours and then compare the open rates, click rates and conversion to see which subject line generated the best results – that is the winner and you can now send the email with that subject line to the rest for your audience.

This is a great way to a get a ‘live’ test on which subject line performance the best at that time. There is of course (like with all aspects of email)  a ‘seasonal’ effect on email results. Seasonal not just being time of year (Holiday time etc) but also time of month (near to pay day) or at the weekends. Therefore running a pre-mailing split test enables you to see which subject line is performing best at the actual time of send. That is why I call it a ‘live’ test.

5.   Large image vs large copy
Best practice tells us you should have a good balance of HTML to image ratio and not have lots of copy in your email marketing message. However is that always true? It depends on your market sector and audience – and who really knows without testing?!

Try testing a more image-focused email vs an email that has lots of copy and see which one gives you the best results.

Remember to not just look at opens and clicks, but try to look at website behaviour and conversion. Conversion doesn’t have to be a purchase; it can be someone completing a form, downloading a whitepaper or requesting a demo. If you don’t track those metrics currently, email me and I’ll show you how.

6. Time of day and day of week testing
Time of day and day of week test testing allows you to see when is the optimal time to send your email. It is good to test; however can also be tricky manage. Why is it good to test? Well not all audiences are the same and people use email in different ways.Some people believe in sending email during the night (whatever time zone you are in) so the email is in the inbox in the morning. Or on a Friday afternoon when everyone has switched off from work. However it entirely depends on what you have to say, what you are selling and your audience – therefore it needs testing for YOU, for your audience and not for what others are doing,

This is what I recommend.

  • Calculate the total list size divide by 24 hours and then send every hour to that amount. For example if your total audience size is 240,000 you would send 10,000 every hour for a 24 hour period.
  • Split your audience into 7 and send at the same over a 7 day period. Wait 7 days and then see which day performed best.
  • Don’t change both variables for this test
  • If you are sending a short time based offer (48 hour online sale) then this type of testing is irrelevant
  • Remember time zones. The west coast of the US is 8 hours behind the UK (and the rest of Europe) so run your tests country specific.

7.   Image and call to action testing
I have jabbered on about testing email creative components that are the subject line, pre-header, From Address and image vs copy but what about the call to action? Testing the call to action is a very cool way to see what is resonating with your audience, driving them to click on the email and actually take action.

Try testing the following:

  • Call to action vs no call to action. Yes sounds weird I know, but people are used to clicking on images therefore you may not actually need a call to action as such
  • Call to action colors
  • Call to action language
  • Call to action as image vs HTML/images so it is viable before images are downloaded. This is called the bullet-proof button.
  • Call to action placement. Top of email vs bottom of email. Left or right of image.

8.   Normal vs Mobile version
It is a fact that more people are reading email more frequently on a mobile device, and therefore you should be taking steps to optimize your email for mobile devices. Using Kickdynamic.com will allow you to track what device your email is being opened on and the duration of time the email is read. This gives you an idea of what your audience is doing. Then I recommended testing.

You can test:

  • Thin version of the email vs normal width (650px)
  • Size of the call to action button
  • Fully optimized email version vs normal version
  • Adaptive email design vs normal email version

Remember to not only think about email metrics but think about what happens next; where and for how long the recipient is visiting your website and if they convert.

What else do you test?