Before you jump into creating variations of your control page, formulate a hypothesis. Hypothesis is an assumption about the occurrence of an event. For example, if you believe that changing your checkout button color from blue to green will improve conversion, then your hypothesis would be something like:
“Green Checkout buttons attract more clicks than blue checkout buttons. According to 2012 Studies on Eye Tracking; Green color has been proven to attract shopper’s attention”
This is a testable hypothesis as you can use the 2012 Studies on Eye tracking to find out variables that prove a higher shopper’s attention. It can be click through rate, time on page, bounce rate or scroll rate.
But a much more effective way to create testable hypothesis is to use analytics data for the control page. In this case, take the page with blue buttons and list the following variable values
1) Click Through Rate (CTR)
2) Time on page
3) Bounce Rate
Now use the most important variable for the experiment to create a testable hypothesis. For this experiment, CTR is the most important one. According to Analytics, Click through Rate is 15% for Blue Checkout button
“Green Checkout buttons will attract more than 25% click through rate compared to 15% for Blue Checkout buttons”
From the start itself, we had shortlisted the web element for this experiment (Button Color). In real-life experiments, you have to run a series of tests to find the web elements that influence conversion in your webpage. Otherwise, you would be guessing and the tests would take considerable number of days to reach statistical significance.
In general, the following web elements influence conversion
1) Headline
2) Copy
3) Button
4) Page Layout
5) Background colour
6) Font
What factors influence conversion?
1) Trust: Always assume that your visitors are sceptical about your offer, your product and your company. What can you do to gain the trust of the visitor? It can be guarantees in your offer, testimonials about your product and information about your company (contact, address and team), that can regain the trust.
2) Relevance: Visitors coming from different traffic source (search, social media and paid network, email) behave differently. However, on reaching the page, every visitor immediately asks one question – “Have I reached the right page?” The headline, copy, navigation and related pages will influence relevance.
3) Distraction: If the web elements in the page are distracting and hinder the visitors from completing their task, then the conversion will go down. Some of the distracting elements are advertisement inside the content that obstructs the reader’s eye path, video auto-play and auto-expanded advertisements.