Email Subject Lines: A Quick Assessment of 7 Split-Tests While working on an email campaign with a client recently, one of the peeps on the phone said this startling thing:

“We saw in several split-tests that short subject lines work better than long ones for us, so that’s why we’re using short.”

My jaw dropped.

Because, for as long as I’ve been in dis biz – and with as many email subject lines as I’ve split-test for myself, past employers and a world of clients – I’d forgotten that some people actually try to learn from their subject line tests. Let me explain…

See, we learn from A/B landing page tests. Headline tests. Button color tests (hew boy). In-app messaging tests.

But don’t most of us just split-test our subject lines so we can be sure we’re sending the best possible candidate to the remainder of our list for that one campaign? The idea of learning from these tests brought back vague memories of trying to do exactly that early in my career but butting heads with an email marketing control-freak (who hoarded email data like this lady hoards puppets). And If you need a definitive guide on how to split test, start here.

Thus, I’ve decided it is high time I get serious about DOING SOMETHING REAL with the results of my email subject line tests. Which brings me to…:

The Winners + Losers for 7 Recently Split-Test Email Subject Lines

With ‘open rate’ being my go-to success metric for email subject lines – which is not to say that subject lines don’t impact metrics like clicks or that subject lines are the only things that influence open rates – here’s how 7 recent email subject line tests for Copyhackers panned out.

Note that these emails were sent in January and February of 2013 and, as mentioned, are for my weekly newsletter (i.e., not direct sales pitches). I’ve ordered them from ‘biggest winner’ to ‘smallest winner’.


heart your home page [Copyhackers]

LA-HOO-ZE-HER (aka Loser)

how to show your home page some love [Copyhackers]



possibly the most overlooked home page opportunity in the universe…?

which home page kills it? a comparative analysis [Copyhackers]


when it comes to email, does size matter? []

no, size doesn’t matter []


increasing paid conversions by 61% on a typical startup site

increasing paid conversions by 61% on a product catalog


how the right tone turned around this small biz []

does the tone of your copy really matter? []


can you write 1 page that speaks to 2 audiences? [Copyhackers]

how to write 1 page for 2 audiences [Copyhackers]


weigh in on 5 headlines… [Copyhackers]

dare to disagree? [Copyhackers]


In another world, I would have formatted the above table so you could see the math. As in, winner’s open rate minus loser’s open rate. But… **grimace and change topic**

Takeaways for Future Email Subject Lines

Judging by the above breakdown of winner vs loser, I would be wise – in the future – to do the following. And you would be wise, dear copy hacker, to assess your own subject line tests accordingly, too.


  • Begin “how to”
  • Suggest un-fun work, such as “comparative analysis”
  • Talk explicitly about product catalogs
  • Are vague
  • Are cheeky with no substance


  • Use as many words as necessary when a subject is intriguing
  • Make explicit reference to my reader’s world: startups
  • Focus on specific keywords, like “home page”, “headlines” or “email”
  • Suggest a case study within

Phrasing subject lines as questions doesn’t seem to pull one way or the other for my list.

It also seems that I can worry less about including my brand name in the subject line. (At least, that seems to be the case when subject lines are >50 characters with spaces.) Given that my From line is “Joanna at Copyhackers”, you’d think this would go without saying. But SO MANY TIMES I’ve seen branded subject lines outperform unbranded – as I discuss here – so I had to see it with my own eyes, for my own list, to believe it.

With that in mind, my subject line test for this week’s Tuesday morning newsletter will be designed around my hypothesis that removing my brand name from my subject lines will not negatively impact open rates. The test itself will use short subject lines – as in, not >50 characters – because short lines simply allow more room for the branding to appear in an inbox. Depending on how this pans out, perhaps a test of subject line lengths is also in order.

I’ll be posting the results of my test here on Tuesday (after my Unbounce webinar, which you’d best be attending, yo!).

In the meantime, here’s a quick ‘n’ easy tweet for your followers… and be sure to tell me your take on subject line testing in the comments below…

As in, do you test your emails? Do you learn from each split-test before writing your next subject line? …If not, take 20 minutes today to make a list like I did above… and sort out what seems to be working – and what doesn’t – for your particular subscriber list.

Happy copy hacking,

As promised, I split-test my subject line for my newsletter on Tuesday. The 2 subject lines and their open rates are:

startup Qs about landing pages + emails [Copyhackers] – 36.4% open

startup Qs about landing pages + emails – 32% open

So I was WRONG about “[Copyhackers]” not mattering. It appears that the presence of my brand name may increase opens. What does that mean for my original question? I’d love to hear what you think it means – so share your thoughts, y’all!