Email Marketing – Testing

So you have decided to get started with email marketing. You chose the system you would use. Set up your list and added a few people to it. And you are ready to send your first newsletter. Or you’ve sent a few newsletters but wondering why you don’t get many clicks, opens and enquiries for special offers you were promoting.

One immediate area we need to look at is whether you’ve done any testing before sending you campaign.

Testing – the basics

With emails it’s good practice to start planning your testing from the start – before you’ve actually finished designing a template. It may be an idea to have several different versions of your content and design so you can test them for effectiveness. What is effectiveness in this case? It depends what you choose to measure as an important parameter for your business:

• click on links matching with Analytics reporting (own site)
• clicks on links for adverts you had included in the newsletter (other sites)
• opens of the newsletter – good one to keep an eye on as without opens you can’t get clicks. However it’s important to note that not opens are recorded properly by email marketing systems. So what we are actually tracking is not the actual number per campaign, by a trend over a number a campaigns – we’d like to see the number climbing up!
• shares of your newsletter – via Forward to a Friend option (if your provider offers it), social media, blogs
• new subscriptions to your newsletter as a result of shares, recommendations of others, archive copy you had available
• responses directly to you with comments or enquiries
• finally, direct business you actually get from the newsletter (sales, enquiries, take up of special offers, free product subscription)

The important point is to test one set of variables at a time as you will know then what changes actually work and what will not. What you can test:

• Scheduling – When you send your newsletter, e.g.morning or evening, day of the week.
• Format – Plain text, HTML newsletter, PDF attached to a plain text cover email, complex HTML design or a simple 1-column text-based design.
• Length – Short with links pointing to full articles and blogs hosted elsewhere or long with full content included.
• Promotion – Different wording,. e.g. Buy one get one free, 50% off, 2 for 1; where in the newsletter the promotion is placed; if it’s related to the rest of the newsletter or something completely different.

How to do testing

Email testing can be easily done in Mailchimp or Campaign Monitor (and most other email/newsletter programmes in one way or another).

The most straightforward way to test your content is to take 10% of your list, send half one version of the email and the other half – a different copy. The one with the best response can then be sent out to the remaining 90% of your list.

What you need to consider first:

• Test your email layout first using automated tool that will prepare screenshots of how your newsletter will look like in different email clients (Outlook, Apple Mail, Lotus Notes) and web browsers (Safari, Firefox, Explorer).
• Test your email title or header first. Then test your content.
• Format and Length are best tested over a number of campaigns, you may also ask your subscribers for direct feedback as to what they prefer and why.
• Scheduling testing can be done over a number of campaigns, when different parts of your list receive the same email newsletter on different days or times.
• In order to achieve all this you need to allow sufficient time for testing (so not rush your newsletter through in the last moment) and have a flexible, well setup and well managed database of subscribers.

How often should you email?

With the proviso that if you have told your subscribers how often you will be sending them emails, don’t break your word. If you have not stated this in the beginning, when you have something to say, email them. You can tell by checking your analytics whether the campaign is successful or not. Successful open rates vary but a 5% conversion rate should be possible, if all things are in place within your email. (i.e. links that are active with a strong call to action).

If you find that your monthly newsletter gets longer and longer with each campaign – break it down and send 2 smaller campaigns every 2 weeks. See if this will improve your effecitveness as some readers may not have time to read through a long newsletter, even if you are only including snippets of your articles. Another point to consider – do you actually need to send all of it? Is it relevant and useful to your whole list? Maybe it’s worth breaking it down into subject-matter newsletters and send to only groups of people on your list. Your newsletter can then still be monthly, but only contain information relevant to that particular group of subscribers.

Try different times to send your newsletter and consider the type of people your email is being sent to. Businesses that work 9-5 may not be happy staring at a full email box on Monday morning. What would they do? Get rid of as many as possible as quickly as possible – so your carefully crafted newsletter will just get deleted. On the other hand, some customers may be reading email on a Sunday before the crush of the working week and be more at ease with a weekend email. Or you could be offering something that’s relevant more to a weekend (discounts for children’s activities) rather than work days (business printing information).

Now it’s over to you. Have you tried testing your newsletters? Did it work, what was your experience, did it improve email’s effectiveness? Please share your thoughts below and don’t forget to Tweet or Like the post