One of the great things about email marketing is that you can test and measure results quickly. Sometimes within minutes of sending an email campaign, you can see whether it is having an impact or not.
A recent study from Dot Digital* showed that 52% of marketers surveyed either didn’t conduct any testing, only did basic testing or had no competence in testing email campaigns.
Email testing – It doesn’t need to be complex
Testing your campaign doesn’t need to be a difficult process, it can be as simple as testing one subject line against another. A simple A/B split test can provide useful information that can then be used in the next campaign. Most email service providers (ESP) make this really simple to do.
Just by testing the subject line you could notice a significant difference in open rates, and learn what works and also what doesn’t to your target audience.
What should you be testing?
Subject lines: An important part of any email campaign, it has a huge impact on whether your email will actually get opened or not. If you can segment your list then you can send specific subject lines to certain segments.
Offers – Price testing, discount codes, free downloads. Price testing is a great way to quickly and easily test pricing for new products or changing the pricing of existing ones.
Calls to action – Really important this one, it can make the difference between someone clicking through from your email or not. You can test text versus image call to actions – text links as opposed to banners. Where in the email do you place the links? (I would always recommend adding one high up in the email, above the ‘fold’ of the preview pane) and what does the call to actions say?
Length of email – Short copy versus long copy. It would be easy to assume that you need to keep copy short, people don’t spend long reading their emails, preferring to scan instead.
However, depending on the type of email you send, long copy may well suit your audience. If you have very long emails you could try breaking it down into ‘snippets’ and adding links to the full article hosted on your website.
Layout – Fonts, images, the position of copy or ads can all be tested. Does your email reflect your brand? Is the design of your email consistent with that of your other marketing? Does it need to be? What happens if you do something completely different?
Time, Day and Frequency – A common question in email marketing circles used to be ‘when is the best time to send an email?’ This was before everyone was connected 24/7 via their smartphones.
However, there may still be some good reasons for sending at specific times. Especially in B2B where your audience is more likely to take action during office hours.
What time of day do you send? first thing in the morning so it’s in the inbox for 8.30 or later in the evening when people are perhaps catching up on email after the kids have gone to bed and the washing up has been done? Do people check your emails on their smartphone and take action later?
It’s worth testing the frequency of the emails you send. It’s quite common to only send once a week or even once a month. What happens if you send two emails per week?
When increasing frequency, don’t just look at open rate. Look at the actual number of opens from all campaigns. You’re likely to see a dip in the percentage open rate. However, the total number of unique opens is likely to increase.
You’ll need to make sure you’ve got sufficient content for the extra emails, and this will need some planning in advance.
What happens if you send emails at a weekend? Does timing matter when you send these?
Who is your email from – This has an effect on your reputation too so be sure to get this one right.
Do your emails come from your company name or brand or from an individual person within the company? What happens when you test one against the other? Does having the email come from a human rather than a company name have a positive or negative effect on open rates?
How to test your email marketing
Multivariate test – You may have a large database, so send the test to a small segment of your list and roll out the winning test to the rest of your list. You can do this for subject lines and creative.
Depending on the size of your database, you may be able to roll out the winner the same day, or more common is to roll out the winning variable the next day.
A/B split test – or A/B/C testing, depending on the variable. For example, you could test 3 subject lines to see which one gets the better open rate. A/B price test, £5.99 Vs £7.99
Have a control group and test against it. For example, you could exclude 10% from getting your abandon cart email and test whether sending the email makes a difference to conversion.
IMPORTANT – Remember to only test one element at a time. Performing more than one test will skew the results and make it very difficult to determine what caused the change in results.
If your list is small, you will likely need to perform several tests to get results. For example, if you doing a subject test where you test discount in the subject line versus no discount in the subject line, for example, you may need to send several variations before you have enough data from which to draw conclusions from.