<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=105052673235362&amp;ev=PageView&amp;noscript=1">

A/B Testing for Sales Emails: What Should You Test?

Posted by Casey Lewis on Monday, December 22, 2014

Topics: Sales Process

sales-emails-ab-testingThere’s a harsh reality you have to face when it comes to sales emails: no one wants to receive them. When you checked your email this morning, you probably received at least one sales email, and you probably ignored it. When it comes to sales, Americans are far past the saturation point. They don’t want to hear your sales pitch. Businesses place signs banning solicitors in front of their stores because they don’t want their customers to get annoyed. We use our DVRs to skip past commercials because we’re just not interested. We may be a materialistic, consumer-driven society, but we still don’t want to be sold to – we’ll handle our own purchases, thank you very much.

Unfortunately, in marketing and business, sales emails are a necessary evil. If you don’t send sales emails, you’re just leaving money on the table. Because as much as we may hate them, sales emails still work. For every $1 spent on email marketing, there is an average return of $44.25. Could you use a return of 44.25-to-1 on a business expense? I thought so.

So what do we do about this conundrum? How do we effectively implement email marketing (which clearly works) when so many people hate it?

Testing. Yes, loads and loads of testing. The A/B test is nearly as old as the sales email itself, because it provides us with valuable information and the ability to adapt to our subscribers. But then people wonder, what should I test? My favorite answer: absolutely everything. That’s a slight exaggeration, but it is hard to test too much. No matter what you change between each version, it has the potential to provide valuable insight as to what works and what doesn’t.

But if you’re interested in specifics, here are some more in-depth ideas on what to test:

Subject lines

This is the most popular test idea. Don’t change the content of the email, just change the subject line for half of your subscribers. Traditionally, teaser subject lines seem to work the best (e.g. “The secret that will change your marketing forever”), but not all recipients are the same – yours might respond to something different. 

Content among similar recipients

One thing to keep in mind about tests: the test recipients need to be similar. That way, you know that the only difference between the recipient groups was your email changes. When it comes to changing content, this is extremely important.

Let’s say you want to send a sales email focused on the benefits of one of your products. Write one version of the email highlighting certain benefits, then write an entirely new version from scratch, making sure you try a different approach. You can highlight different benefits, write it in a different tone (more educational compared to more casual, for example), or even experiment with different lengths. Just make sure the bodies of the two emails are very different. This can help you understand what approach your recipients are more likely to engage with.

Content offers

In inbound marketing, content offers are a crucial part of the equation. They really help you turn site visitors into leads. You should also occasionally send out sales emails with content offers to aid the lead-generation process. Send emails with different content offers and see which ones get the best response. If you have a large enough recipient group, you might consider sending out three different content offers and see which works best.

Important tip: DON’T Cross-test

What do I mean by cross-testing? Don’t change everything between emails. For example, if you’re going to change the subject line between two groups, leave the content the same. Otherwise, you won’t know what caused any difference between opens and click-throughs. Was it the subject line? The body? The content offer? Change only one thing between groups to help you gain insight, otherwise you’ll have confusing results.

New Call-to-action

Comments