The internet provides us with an endless number of ways to test and perfect our marketing communications. Like mad scientists, we can experiment with our website using Google’s Website Optimizer, continuously tweaking it to better. Tim Ferris even talks about testing product names and price points using eBay listings and AdWord campaigns.
Something that I’ve been playing around with is using Twitter and the url shortening service, Bit.ly, to test the effectiveness of headlines. It’s a quick way to find out which phrasings work and which don’t.
Here how you can gain quick insights to perfect your headlines:
1. Pick a URL Shortener
First pick a url shortening service that will let you create two or more short urls for your link and that collects click data on your urls. I like Bit.ly because it is easy to use and it displays data in an clean format that takes no time to digest. (Plus it is the most common url shortening service.)
2. Select Your Content
Now find the content that you want to test headlines for. In this example, I picked an article on the Dachis Group’s Social Business Design.
3. Create Two Different Short URLs to the Content
Copy the url from the content page you want test headlines around. Create two different short urls linking to this page.
4. Pair Each Short URL with a Different Headline
Pair each of the short urls to a different headline that you want to test. (Using Bit.ly, this is when I rename the link titles so that I know which headline I used for each link.)
5. Tweet the Headlines Followed by the Paired Short URL
Tweet the first headline you want to test with the url to the content page. Then wait a few minutes and tweet the second headline and url.
6. Track the Results
Using your url shortening service, track the number of clicks on each headline. If you are using Bit.ly, the top number is how many clicks there have been and the bottom number is how many clicks there have been total, through Bit.ly, to the long url. That is why the bottom number for both links is the same, because they are both linking to the same page.
So from this we can see that “Social Business Design” is a much more effective headline hook than “Improving Value Exchange.”
7. Repeat to Find the Perfect Headline
This is where you start the process over. Go back to step 1 pairing up the most clicked headline from the first test with a new headline variation. As you do multiple tests make minor variations in the headlines in order to better understand the results. (e.g., instead of using the word “design” you might try using the word “blueprint.”) Do as many tests as you want, until you are satisfied with your headline. Depending on your timeline and what you are testing for will determine the length of time you want to spend testing.
I should point out that there are some limitations to this testing method:
- For this to be of any value, you need to have a history of participation and a following of enough size to accurately make a comparison.
- There is no way of knowing whether or not users clicked on both links.
- Participants on Twitter won’t always match up to the audience you are crafting a headline for.
That’s it! A quick way to experiment and perfect your headlines.
That is a great post on how to test your headlines, but here is the concern:
As you said “Participants on Twitter won’t always match up to the audience you are crafting a headline for.” People are different, and they get attracted to different things. What some of your followers might find as a headline and link worthy of clicking, might not seem so for the rest. And since in Twitter, you have no control over the type of followers (beyond to who you decide to block, and you surely won't do that), you might be left with a large number of followers but only a few who really are your targeted audience and would pretty much click on most (if not all) the links you tweet because they know they'd find value there.
And as you said: “There is no way of knowing whether or not users clicked on both links.”, your loyal followers would most likely click on both links, only to find out that it is the same content that they have already read. To me, that might cause a negative reaction, like they were being tricked only for the blogger to increase readership. I say the best solution to that is to leave an interval of time before posting the second link, it would mostly be more accurate if you post it in 12 hours, why? because there is a less chance that a follower who was present to click on your link the first time be online after 12 hours to see it and click on it the second time.
And I'm sorry that my comment seems more like a blog post than a comment!
As you've pointed out, this isn't perfect. It's just one more way to test.
The amount of time in between both links is an important part of the test. I should have specified a set amount of time in between tests. Firing back to back tweets with the same link is a good way to tick people off and lose your following. As a rule of thumb, I usually send out the tweets 6 hours apart.
As always, thanks for your comments.