Optimizing Through Testing

August 08, 2017 / Marko Mihin

Introduction

If you use Sitecore, it isn't uncommon to have a couple different ideas on how to display information to people visiting your site. With multivariate testing, you can try out multiple ideas and pick whichever one performs best,  based on real data.  Setting up two versions of every component on every page might be time consuming, but the effort is worth the reward if you are able to optimize your conversions and take out the guesswork.

 

By testing and changing certain parts of your site you can achieve

  • Higher conversion rates
  • Increased overall traffic
  • Improved User experience
  • Reduced time to purchase

 

Summary of Key Features

  • Simulate visits to your website to see how different personas interact and observe the effect on ‘Engagement Value’ or level of interest in a given marketing asset
  • Detailed view on each experience tested in order to analyze the Value, triggered goal, clicked links and other usage metrics
  • Automatically test content changes, personalization rules, goals, events at component or page level
  • Gain an overview of all optimization efforts via leaderboards that report on individual performance according to four KPIs (score, activity, prediction, and effect)
  • Different user roles available; basic testing for a content author versus robust tools for an optimization expert
  • Built-in machine learning analyzes the data and provides recommendations
  • Use customer insights gained from Sitecore Optimization to support marketing programs across web, email, social, mobile, commerce, and print channels

 

Workflow

The test cycle is a simple process to follow:

  1. Create hypothesis and test it
  2. Analyze the results
  3. Optimize your website

test_cycle.PNG

 

What can we test?

  • Page Level Testing - Tests two or more pages, or versions of pages, against one another for the best outcome
  • Component Testing - Measures value provided by differing components
  • Content Testing - Measures value provided by differences in text, imagery, video etc.
  • Testing one of these "variables" against another one is called an A/B Test.

 

Using A/B and Multivariate Testing

A/B tests are usually performed to determine the better of two content variations.

A/B split testing is a method of marketing testing by which a baseline control sample is compared to a variety of single-variable test samples in order to improve response rates. Simply said it test only one section with multiple variables.

Multivariate testing is a technique of testing where multiple a/b tests are performed on one page at the same time. You can add more than one component to a multivariate test so that you can decide which combination of controls works best on a page. For testing combinations, you need to be aware of your traffic. The test should continue to run until the amount of data collected is statistically significant. Simply said, multivariate testing test multiple sections with multiple variables.

A/B testing allows us to see which content works best in which context, we can then go ahead and make use of personalization. The two compliment each other perfectly

 

The Experience Optimization app

The last thing to mention is the Experience Optimization app, where you can create tests. It has been created to manage all tests and their outcomes in one tool fully integrated into Sitecore. From the Experience Optimization Tool you can view Tests in all states; Draft, Active and Completed and once a number of tests have been completed Sitecore will have learned enough to begin displaying Suggested Tests. These Suggested Tests are what Sitecore has determined would be good changes to better achieve the goals, campaigns etc.

 

Contact us to set up a free consultation with one of our marketing subject matter experts.


Head Office Hours

Global Support Hours