This is the second experiment in our “Data Driven Music Marketing” series. In our first experiment, we tested which type of button would lead to the most music sales on your website.
This month we have run an experiment with the goal of increasing mailing list sign ups. Almost every musician we work with gives away some music to encourage a fan to join their mailing list. Whether they offer a full song, a remix or an acoustic version, we wanted to test which Call-to-Action encourages most fans to sign up.
The Test
We use Topspin to power our fan acquisition campaigns, so we used a Topspin Email-For-Download widget as the basis for this experiment.
For example, an email-for-media widget looks like this:
As you can see, there are two main text elements which could influence the fan’s decision on whether or not to sign up:
- The Headline - “Free Download of Two Live Songs”
- The Button - “Download Now”
The Options
In order to keep the experiment manageable, we decided only to test the “Button” text and to keep the “Headline” consistent.
We took proposals from the Topspin Greenrom (a discussion forum) and arrived at the 5 following options:
1. Download Now
2. Get The MP3s
3. Click Here
4. Enter Your Email
5. Download Free MP3s
The Results
In this experiment we measured the conversion rate from views -> new subscribers. It didn’t matter if one button got more clicks or even more emails entered, we only counted a conversion when a fan clicked the button, entered their email address AND clicked the confirmation email.
The results were very impressive and much more varied than we had anticipated. The difference between the worst performing and the best performing buttons was a whopping 69% !
To measure the effectiveness, we first calculated the average conversion rate (which was 1.12% of views lead to a subscription).
The worst performing button was “Enter Your Email”, which resulted in 24% fewer sign ups than the average.
The best performing button was “Click Here”, which resulted in an impressive 28% more sign ups than the average.
Conclusion
This experiment, which was run on multiple websites over the month of December, had a cumulative 18,362 views and 206 email sign ups.
From this relatively large sample size we are confident to conclude that the average musicians could increase the growth of their mailing list by 25% or more by encouraging fans to “click here” to get a free song.
—####—
Notes: The full raw data is available on a shared spreadsheet, showing all the results and calculations.
The subscription rate is quite low (1.12%) because we didn’t test many widgets in very prominent placing.
Do you have any experiments you’d like us to run? Leave a comment below. We’re looking for more suggestions for our February experiment.
4 Comments
Ahoy there,
Popping in for another examination of yere blog; theres some nice post here since the last time I visited so thanks.
Anywho I was hoping you could confirm that the bar chart is in the order listed at top theres only the best and worst labeled and I’d like to know which the others are as they perform quite closely and I’d just like to know how my gut instinct fares… I would guess’ Download Now’ to come in second but its nice to know for sure.
Cheers and keep well
-Josh
Actually, could you share the spreadsheet too; the link works but its locked.
Cheers
Nothing like a good old call to action, would be interested to see what would happen if you made your button red.
- Chris
@Josh - Apolgoies, my Google docs skills aren’t up to scratch!
Yes, the graph is in the same order as the list. I was raised on excel, so I’m still getting used to making well labeled google graphs!
For the same reason, I didn’t publish the document, but the link has been updated now:
https://spreadsheets.google.com/ccc?key=0AsV6bCrSWDIrdHF0b1RVdXhFSmk1ZEJhYjRYQkhyX1E&hl=en
This should have much more info than the graph for you.
@Chris - we do often run experiments like that, but we don’t publish them here.
The blue was just for example purpose, none of the widgets were actually that colour.
We don’t publish style/imagery based experiments because the results are different for each artist. You can’t really use the results found by another artist and assume they’ll work for you.
For example, would red buttons improve the conversion rate? Red would stand out on our site, so probably… but on a site with a predominantly red colour scheme it would just blend in.
P.S. Any more suggestions let me know, we’re always looking for more elements to test!