By: Rachel Bergman, Head of Email Services/VP Product, Valassis Digital
Published Tuesday, Jan 10, 2017
As a marketer, I have to admit I often dread the idea of coming up with the subject line for an email I’m sending. It’s not that I don’t like a challenge, but I like a challenge that ends with me potentially “winning” or at least with me knowing if I won or lost. So, I have to make a confession – I simply don’t write subject lines anymore. OK, that’s not entirely true. I write plenty of potential subject lines, but I don’t make the decision on which one should be sent out – I let the machines do that. The machines tell me what my chances are of winning before I hit the send button, which I have to tell you is actually quite liberating.
It may be a little known secret, but the explosion of data, and our ability to process and understand it, can actually help us with our email subject lines, and you don’t need to be one of the biggest brands with the deepest pockets to ensure you are using this data to both make your life easier, and help you win. So, here’s what I do:
1. Track performance over time
Keep track of the email subject lines you’ve used in the past. This may seem intuitive, but it’s important. Make sure you have a spreadsheet with your subject lines, the dates you emailed, your target audience, and how they performed. This should help guide your decisions, but not completely chart your course.
2. Search for inspiration
True story, I sign up for a lot of companies’ emails, and so should you. I have them all come into one crazy inbox, and I peruse them for inspiration. I don’t plagiarize, but looking at what similar and/or aspirational brands are doing is just smart. You might also consider signing up for Milled (www.milled.com) which will curate the emails or brands you are interested in following.
3. Start experimenting
There are several ways to do this, of course. I send out a lot of localized email, so it would be difficult for me to use a product like Persado (www.persado.com). It takes the base subject line you write, creates 16-32 versions of it to test in-market with your subscribers, then the machine generates the “winning” subject line. The reason this would be difficult is that you need a large enough sample set of data to make it work – tests need to be statistically significant. This said, if you are a large brand and send out a significant number of emails, I encourage you to at least explore this product – at the very least you will learn something.
My tool of choice these days is called Touchstone (www.subjectlinegold.com). To use this tool, I write several subject lines, based on items 1 and 2 above, plug them in and watch the data come back to me within seconds. I’m instructed on which subject lines have the best chance at being delivered, opened and whether the subject line should lead to clicks on the emails themselves. This tool also provides warnings and recommendations about certain words and phrases I’ve used. It analyzes data on how similar subject lines have performed across billions of emails that have already been sent…so I don’t need to actually test them to know the results. Clearly, I’m over-simplifying this, but hopefully you get the point. I should mention that this tool is far from perfect, but it’s a bit of a game-changer if your alternative is just hoping and praying that your email performs (and even if it does perform well by your standards, you won’t know if it performed as best as it possibly could).
In the end, you can read tons of articles that will tell you the length your subject line should be; how many emojis to use or not to use; and whether or not to use personalization, but unless you test, or virtually test (like I talked about above), you run the risk of not being as successful as you could be. And that’s not “winning” in my book.