Use Split URL testing when you want to test multiple versions of your web page hosted on different URLs. This type of test is considered best when you notice several deficiencies in your existing page that might affect conversion. In such cases, it's often easier to build a whole new page for testing, rather than trying to modify the existing one.

For example, say you want to create two different designs of your home page, with two different URLs, and distribute your website traffic among them to test which version performs the best among your audience with the Split URL experiment in PageSense.

Setup your Split URL test

1. Click Split URL Test on the left pane, then click the New Split URL Test button. 



2. Enter a descriptive name for your Split URL test and enter the URL of the original web page you want to create variations for.
You can also enable Redirect with query parameters if you want to pass the query parameters of your original URL to the experiment URLs. This is useful when comparing landing pages, or multiple versions of a site page that has parameters that change dynamically based on users or user inputs. Learn how.

IMPORTANT: You can also run a Split URL test across a group of web pages by enabling the Advanced button. This option also allows you to choose different URL match types and include or exclude web pages from being tracked in a single Split URL experiment. Learn how to create a Split URL test using advanced URL targeting options. Make sure you don't include the redirect destination URL in the URL match conditions. Expand the Test URL(s) tray to verify that the redirect URL is not included in the match conditions.

3. Enter the name and URL of the other page you want to include in the experiment on the VARIATIONS tab and then click Add. You can test multiple pages by clicking the Add button. 

For example, you want to test two different design versions of your blog page to see which one gets more clicks.

4. Click Next.

Add conversion goals

1. In the GOALS page, click + New Goal button to add a new goal to your variation page.



2. You can choose any of the goal types from the list of options on the left pane:

• Element Clicks: Tracks conversions when a visitor clicks on a specific element on your web page. Learn more about creating an element clicks goal.
  

• Link Clicks: Tracks conversions when a visitor clicks on a particular link on your web page. Learn more about creating a link clicks goal.
  

• Page Views: Tracks conversions when a visitor visits a specified web page URL. Learn more about creating a page views goal.
  

• Revenue: Tracks conversions for purchases made and revenue generated per visitor on each variation. Learn more about setting a revenue goal.
  

• Time Spent on Page: Tracks conversions based on the average time spent by visitors on your web page. Learn more about creating a time spent goal.
  

• Engagement: Tracks conversions when a visitor clicks on any of the elements or links on a web page. Learn more about creating an engagement goal.
  

• Custom Event: Tracks conversions when a visitor achieves a specific event or set of events on your web page. Learn more about creating a custom event goal.
  
  
  

3. Click Save and then Next in the top-right corner to save the goals for the chosen variation. 

Select your target audience segment

1. Use the Participation slider on the AUDIENCE page to set the percentage of your audience to participate in your Split URL experiment.

For example, you have designed two new versions of your home page for your blog site and you have a total of 10,000 people visiting your home page every month. You want to test the original and variation 1 for 50% of the audience who land on your home page. You can decrease the participation percentage to 50% on the slider before starting your test and increase the participation percentage later as required.  

Note: 

1. The percentage of the included users are randomly assigned to either the original or the variants in your Split URL experiment. 
   
2. By default, 100% of users who meet your targeting requirements will be included in the Split URL test. It is better to start off with a smaller audience when you are testing for the first time.
   

2. Choose a specific group of visitors you want to target this variation to based on their behavior, location, devices used, and other parameters in theTarget Audience section. 

PageSense offers three categories: All Visitors (chosen by default), Preset Segments, and Custom Segments. If your requirement doesn't match any of the audience types, click the +Custom Audience button.

For example, if you have targeted an experiment for desktop visitors only, your original and variation pages will be shown to users who access the page on their desktop devices based on the traffic percentage set for each variation. Learn about the different types of audience segments in Zoho PageSense.

3. Click Next to move on to the CONFIGURATION tab of the experiment.

Configure your testing options (experiment method, time duration, etc.)  

For example, you decide to run the split experiment for 80% of visitors (chosen in the Participation section of the AUDIENCE tab), then you segment those visitors iso that 50% see the original site and 50% see the variation page. Now, of the total visitors who come to your website, 20% will not see the experiment at all and the remaining 80% will be split as a 50/50 ratio between the original and variation pages. Once after the test period is over, the winner is determined based on the allocated traffic to the respective versions.

Info: Traffic distribution expresses a probability that visitors will be placed into a particular variation. If you have two variations and the traffic is split equally, each new visitor has a 33% chance of being shown each variation or the original. However, the variation's actual number of visitors shown on the REPORTS page probably won't exactly match the 33% probability, but over time, the number of visitors in each variation should approach this number. 

2. Choose your preferred statistical method to run in the Set Experiment Method section and evaluate your Split URL test results. Zoho PageSense provides two types of experiment methods:

1. Bayesian method: This method calculates the statistical likelihood of A performing better than B.
    
   
   

2. Frequentist method: This statistical method only tells you if A will beat B, but not the probability of A performing better than B. 
   

• Quick Trends: This method is ideal for testing non-revenue-related goals such as quick headline tests or CTA clicks. It provides quick results in a shorter testing time and gives a statistical significance of 90%. 
  

• Optimal: This method is ideal for all kinds of experiments and provides fairly accurate results in a moderate testing time. It gives a statistical significance of 95%.
  

• High Accuracy: This is best for critical experiments that affect your revenue directly and provides highly accurate results with a longer testing time. It gives a statistical significance greater than 99%. 
  
  

  
  

3. Choose when you want to trigger the variation to your audience in the Experiment Activation Mode section as follows:

3. Activate when the page loads: This is the default option that will trigger the experiment as soon as the page loads. 
   
   

4. Activate manually: This option will trigger the experiment when a snippet code that you place directly onto your site or within your app is executed. PageSense provides you with a sample code and you will need to replace the snippet code with the custom code that will activate your experiment according to your specific requirements.
   

   For example, you want to show your web page with new products and promotions when the user tries to close the page. Then you should add your custom code in the close function of your web page to trigger the variation page that asks customers to check out the extra discounts offered to them. 
   
   
   

   
   

5. Activate on satisfying a condition: This option is used to add a custom JavaScript code that returns a Boolean True or False value based on the visitor's behavior on your web page. If the Javascript code returns True, your experiment will be triggered automatically and show the experiment page. Note that the Boolean condition will be evaluated and monitored every 50 milliseconds on your web page. 
   

   

   For example, you want to test a variation page that shows a special newsletter pop-up to subscribers who scroll down half way through your blog post. 

   
   

4. You can also enable or disable the following options in your experiment: 

3. Generate heatmap reports: Enable the Heatmap button to see visitor engagement across different sections of your variation and original pages.
   
   
   

4. Schedule an experiment: Use the calendar to choose a specific date to launch your experiment on. 
   
   

5. Connect with third-party tools: This option allows you to integrate your PageSense projects with other third-party applications like Google Analytics and Intercome to access and read your Split URL test reports within your preferred tool. 
   
   

5. Click Next to be taken to the SUMMARY tab.

Launch your Split URL test

The SUMMARY page gives you an overview of your experiment's configuration options such as the number of variations created, goals added, targeted audience segments, and other settings you have made.

1. Click Launch in the top-right corner to launch your experiment.



2. Once your experiment is launched, a new tab called REPORTS will appear next to the SUMMARY tab. 
The EDITOR, GOALS, VARIATION, and CONFIGURATION tabs will be hidden when the experiment is running. If you wish to make changes to your experiment, click the Pause button to make the required edits and then click Relaunch to run the experiment again.