Is your Landing page really doing good? can it be better?
You must have heard someone saying, “wow, that’s an awesome Landing Page.” How awesome? Are you sure that the conversion rate you are getting on your current Landing page is the best? Are you carrying any A/B testing on it? No? Then I doubt how do you say that.
On Internet, the same user changes his behavior from one web site to other. I might look for girls on Facebook with the purpose of dating, but i won’t do that on Linked in. I’ll become a serious guy in the latter, even if the same girl is there on both the sites. You can either change your site according to the user behavior or change the user behavior as per your site. The latter is very challenging but gives rewards.
e.g. On facebook, every 15 days you see a change, may be profile updates are changed in to news feeds. you don’t question it and change the way you surf facebook and remember it next time. ofcourse the changes are made keeping user in mind, but they tend you to change your behavior.
e.g. while on gmail, they knew that people have a lot of junk mails which don’t go to Spam folder, so they introduced Priority Inbox.
In both the cases, users are happy. It depends on which way you want to chose.
Any way you chose, but to see whether users are behaving the way you want them to, you’ve got to test your pages. Test your pages with variations.
How it works?
in A/B tests, 2 versions of the same page are tested among 2 groups of users at the same time. All your users are split in two on random basis and are served the page they are selected to see. When they perform an action and land on a conversion page, it forms a conversion. It’s necessary to have only one conversion page for all version you are testing. A conversion page is normally a Thank you page for authenticating the action just performed. A conversion here can be just a click, or download, or a form fill.
How many pages can I execute A/B test at once?
Test 2 pages or 10, up to you. The thumb rule is “the lesser pages you test, the quicker results you get”. Sites with huge traffic might see results in a few days.
These are the variations you should test:
Call to actions
Call to action can be a Text or a Button. Anything that tells a user to do something is a call to action. These are the things you should test in call to actions:
– Text – Test text variations.
e.g., “Buy now” vs “Add to cart”, “Free Download” vs “Get it free”, “Fill up this form” vs “Register here”
– Buttons colors
Some colors attract users more. But it depends on the background to foreground combination. Also depends on the Target Audience. Red button on a Travel site might not do well. But it might do great on a Music site.
e.g, red vs yellow
– Buttons and text – See if Buttons or Text is performing better.
A/B Testing for Sign up Forms
Form is the second most important thing on a Landing Page. It gets you more Form fills just a small change.
– Position – Change the position. Test Left vs Right, Horizontal vs vertical.
– Split – If you have a long form, Try splitting a log form in to 2-3 steps and justify each step. Then test Normal vs Split.
– Size – Drop some fields you think are not so important and can keep users from filling in. Try removing any “not so important fields”. Test Long vs Short.
A/B Testing for Header text
Header does a very good job grabbing users attention. Always keep the header simple, descriptive and as short as possible.
Test different text in the header. e.g. “Great deals on Movie tickets” vs “Cheapest movie tickets in town”, “Great articles on Google analytics” vs “Google analytics at its best”
A/B Tests for Images / Banners
Images are very easy to engage users. Sometimes, it helps convert more with a minor change in communication or call to action.
– Image size – it might happen that a smaller image with some breathing space might increase clicks by a good number. Sometimes, a bigger image can bring more users. Test sizes.
e.g., Small vs Big, Horizontal vs Vertical,
– Communication / Call to action – The call to action can be tested same as described above. The communication can be tested for variations of influential lines.
e.g., “Tack visits easily” vs “The best way to track visits”, “Great shirts at great prices” vs “Best deals on shirts in town”
It’s not necessary to test all the features of an object. Test one at a time. For testing features of call to action, test Text first, then test Button colors, and then Buttons and Text.
What types of Pages can be tested?
Any page where some action from the user is required. Here are examples:
A download page for software, articles, books, videos or photos.
What content to executing A/B test on :
Call to actions
Position of the button
Images if any. if there’s no image, try putting one.
A/B test on Lead generation page
Sign up page with a form for a website or free trial for a product. Or A lead generation page for a product trial or a bank account.
What to test:
Header of the form, Position and size
Page for a particular product display. Or a multiple product page. e.g., A T-shirt or an MP3 player.
What to test:
Call to action
Likewise, there can be many other pages and Variation you can test. But most important, Please test.
A/B Testing Tools
The most widely used tool is Google web optimizer. It’s free. There are other Free and Paid tools available, but when Google does the best job, why go anywhere else. However, for paid tools you can learn our review on new A/b Test tools.
We at tatvic are specialized in Conversion rate optimization projects and A/B test is our love. And to spread the love, we are offering 3 FREE A/B tests a month. You can be the lucky one to grab one. Doesn’t matter even if you test your pages, there’s no harm in trying for free. If you have never test your pages, then you have no right to wait,
Latest posts by Kamaksh (see all)
- E-commerce Recommender Series: Where to Place Your Product Recommendation? - October 20, 2014
- Infographic: Call-to-action – Guidelines and practices to ensure more clicks - August 27, 2012
- Infographic: An E-Commerce checkout process your customers will love - April 16, 2012