The Art of the Subject Line

Last year, in a talk about online writing that I gave to a group of talented undergraduate writers at the University of Southern California, I argued that the most important genre of writing for them to master is the email subject line. I think a lot about subject lines: what makes them good and bad, effective and ineffective, but one thing that has long puzzled me is how little attention many companies pay to the lowly subject line.

For perspective, I turned to Nick Usborne. Nick is a leading authority on the subject of writing for the web, and he wrote "Net Words," a bible for both copywriters and writers of content online. Recently, Nick joined the talented folks at as a senior analyst. This week, he spared some time to chat about subject lines.
iMedia: Back in the day, I was the editor for all things digital at EarthLink, and in my experience the subject line was always the last thing on the agenda for whatever agency was working on whatever marketing email -- particularly when it came to HTML or rich media emails. This worried me, particularly since the subject line is often a make-or-break component. Do you have a sense of a) whether or not this is as pervasive a phenomenon as I think it is, and b) if so, why it might be?

Nick Usborne: I agree. Many companies pay very little attention to the subject lines in their emails and newsletters. Evidence of this lack of attention can be found in the very basic mistakes so many companies and organizations make.

Why? I can think of five reasons.

1. Lack of understanding.

Many companies pay plenty of attention to the body of the message, and far too little to the quality of the message in the subject line.

To illustrate this point, no offline direct mail agency would rush or hurry the writing of the text on the envelope of a direct mail promotion. Direct marketers understand that the envelope copy is crucial to the success of the whole mailing.

Companies online are yet to understand that subject line copy is just as important to the success of an email or newsletter.

2. Creative pride.

Writing subject lines is all too often seen as boring, uninteresting work. Such a short snippet of text may be seen as "beneath" the talents of the writers involved. Misplaced pride.

3. Lack of process.

Creative and production groups are often under incredible pressure to deliver the next email or newsletter. Because writing and testing the subject line is not carved in stone as a required step in the process, it simply falls between the cracks. A line is written quickly, without enough thought as to its potential impact on open rates and conversion rates.

4. Laziness.

Some writers know how important the subject line is. They also know that others in their organization may not. So they can get away with rushing the job. Pure laziness.

5. Lack of testing.

The first day you test different subject lines and see the results, you'll never be complacent again. When you do the math and see how much money you are leaving on the table with the second best subject line, compared to the best...that's when you start taking subject lines seriously.

iMedia: So what are the "very basic mistakes" that many companies make when it comes to subject lines?

Usborne: The most basic mistakes I see are as follows:

1. Using words that cause the email to be filtered at the gateway, ISP, server or individual account holder level. Do you sell "excess inventory"? Be careful: use of the word "excess" may get you filtered. Do you want to point to a "hot" topic? The same problem exists with the word "hot" and many, many others.

2. Being too promotional. Yes, one can be too promotional. The problem is that your line could end up "sounding" like a spam headline. When spam isn't filtered by technology, it's filtered by the human brain. And the human brain filters based on both the words it sees and the tone of the message. If you are writing subject lines, then you should study spam subjects lines so that you can get a feel for what your own lines shouldn't sound like.

3. Failure to be interesting. Boring subject lines work only when you have a very, very strong relationship with your subscribers. For the rest of us, the line has to touch on something that is important to the reader. You have very few words to work with, so you need to make a careful study of which words acts as triggers for your audience. How do you know which these words are? Through research and testing.

Not sure how to test for trigger words? Use Google AdWords as a research tool. Test different headings for otherwise identical ads, find the top-performing words and phrases and then try them in your subject lines.
iMedia:  On that basic level, how many characters do you usually dedicate to the meat of the subject line? I usually want the meat of the subject line (the product, promotion or company) to be explicit within 60 characters, but with bigger and wider screens becoming more prevalent, I wonder if I'm dating myself with that number.

Usborne: Sixty characters sounds like plenty to me. If you can't get an "open" within the first fifty characters, you're probably not being careful enough with the words you choose.
iMedia:  Recently, we've been experimenting with our own subject lines here at iMedia Connection. We've gone from every subject line being the same (the title of the newsletter plus the date) to unique subject lines. In a few weeks we'll start A/B testing, but right now we're alternating between single-topic subject lines (in which we pick the story we think will interest the greatest number of our readers) and more inclusive, multi-topic subject lines.

I haven't seen the tracking numbers yet, and so I don't know if there will a) be any particular lift or decline in open rate from a unique subject line, and b) don't know if a single topic subject line will be better than a multi-topic one. Do you have any predictions? Regardless of your willingness to play Karnac the Great, do you have a general brief about single versus multi topic?

Usborne: It has been a long time since I tried predicting winners among subject lines. I used to work with a company which sent out over two million newsletters a month. Two days before the publication date we would test between five and seven subject lines, sending each to about five thousand subscribers. Whichever line won would become the line we would use for the balance of the two million.

Here's what I learned. You never, ever know. Over a period of two years, other than by pure chance, I don't think I ever predicted the winning line based on my experience.

As far as I am concerned, the only way to know which is the best subject line, or even the best approach, is to test a range of options every single time you create a new email or newsletter.
iMedia: Earlier you said, "The first day you test different subject lines and see the results, you'll never be complacent again. When you do the math and see how much money you are leaving on the table with the second best subject line, compared to the best...that's when you start taking subject lines seriously." Can you give us a sense of scale by this? How much money? Are you talking about A/B testing or something more complicated?"

Usborne: In testing subject lines I have seen one line outperform another by close to fifty percent. (In that instance, the subject lines were very, very close in both tone and the words used, which only goes to show how important testing is.)

So, let's do some simple math here on an email to a modest list and a reasonable priced service.

Send out an email to 100,000 people. And let's say you want them to sign up for a subscription service priced at $34.95 a year.

If you get a two percent conversion rate, your revenue from that email will be $69,900.

If you get a four percent conversion, as a result of a better subject line and higher open rate, you'll gross $139,800.

Now let's assume that each subscriber stays with you for an average of three years.

Now think about sending this email, or one like it, every month.

And now think about what happens when you test five subject line options and find that one of them outperforms the others not by one percent but by fifty percent.

What are you looking at after three years? You're looking at a difference in revenue that can add up to hundreds of thousands, or even millions of dollars.

This is worth keeping in mind the next time you just scribble out your best guess at a good subject line.

Nick Usborne is a Senior Analyst with and an editor of the Marketing Experiments Journal. The Marketing Experiments Journal is focused on just one task: to find out what really works online. A free subscription to the Journal brings you their latest research findings, twice a month.