For a couple of years I have been the author of the UK Direct Marketing Association's Email Marketing Benchmaking Report. It's aims are fairly modest - to share average campaign metrics from about 26 of the email service providers working on UK campaigns and provide some insight into trends and issues. If truth be told, I have a bit of a problem with averages, and here is why...
- The report covers BtoB and BtoC
- It includes service and product marketing
- It includes all industry sectors
- It includes best, recent customers and worst, lapsed customer
Maybe the slide below explains it better...
In short, averages conceal the really interesting behaviour of segments that we as digital direct marketers shold be investigating. We walk away content that our "average click through rate" is not so bad after all. But what if your email file contains less than the "average" number of active customers who tend to respond better than prospects? You have no idea if you are doing better or worse than average, and that's not very good.
So this brings me onto research that Alchemy Worx has been promoting recently about the relationship of email subject lines and success:
“In summary our findings show that shorter subject lines optimise open rates, while longer subject lines optimise both click and click-to open rates. We were also surprised to identify a “dead” zone! Subject lines of between 60 and 70 characters (6-10 words), optimise neither the open rate or click to open rates.”
This kind of research, in my humble opinion, gives marketers false hope. It oversimplifies the complex chemistry that determines email marketing success and suggests that a specific length of subject line will deliver results. Alchemy Worx then proceed to reel off a lorry-load of qualifiers about the proposition itself as to make their sweeping generalisations redundant.
So, the report just looks at 1 of many factors in un-weighted samples and comes up with some "rule of thumb". This is not necessarily good direct marketing because:
- It deals in big, bad, lumpy averages that hide the best/worst performing segments
- It treats active, loyal customers and old prospects as equal in proposition responsiveness
- It did not test long versus short, so we have no control to gauge their impact
- It does not mention different From Fields, or how well the "preview pane" area was exploited - critical "open rate" influencers.
Out of respect to direct marketers everywhere we should qualify and re-phrase how to use this report
“In summary our findings show that some campaigns with shorter subject lines got high open rates, while longer subject lines got a high click and click-to open rates. Because we never ISOLATED subject lines as a variable and TESTED long versus short we cannot say with any confidence that the extra opens and clicks were down purely to the subject line length."
So here's my advice based on 10 years of email marketing:
- Go and do your own tests - your products, brand, customers are unique and special
- Segment your file to identify the extremes, not the averages
- Test short and long subject lines, but use proper "control" groups
- Test different From fields - they have a huge impact on open and click through
- Don't be too prescriptive - use the right words to persuade regardless of length.
- Be relevant and engaging - show people you know and value them
And finally, now I've got my angry hat on, here are 5 subject lines that are in the Alchemey Worx "dead zone" (60-70 characters and 6-10 words) that I would definitely have opened and clicked...more to do with my interests in life than the word count?
- Sleep better - stop your springer spaniel barking at night
- Crystal Palace win race to sign Ronaldo, Lampard and Crouch
- Welsh Rugby Union offers free tickets to email marketers
- 20 fun family activities when its pouring with rain in West Wales
- Why averages are a bad bad thing in digital marketing