Introduction to web experiments in Sitecore Personalize

Abstract

Provides an overview of web experiments, which facilitate personalization on web pages, without relying solely on product teams (Sitecore Personalize).

Designed specifically for marketers, a web experiment facilitates personalization on web pages, without relying solely on product teams. A web experiment is an offer, next best action, operational message, or any other customer experiment that can run on the web or be deployed into a web-based application.

Sitecore Personalize enables you to test and preview a web experiment and see it display on your organization’s website.

Sitecore Personalize can provide you with web templates as a code-free alternative to web personalization. Or, if you are familiar with HTML, CSS, and JavaScript, you can create these yourself.

When the web experiment is live, you can monitor its performance through operational metrics. If it is in an A/B test, you can see how it is performing against other variants.

A web experiment is effectively an A/B test. You can view analytics to see which variant is leading or if Sitecore Personalize has declared a winner. The goals you set for the web experiment directly impact the metrics you have available in analytics.

You can run experiments on the following UI elements:

  • Design layout.

  • Images - including the wow-factor, placement, number, and style.

  • Headlines - varying the content, length, size, font, and so on.

  • Company branding.

  • Text - changing the content, style, font, size, and placement.

  • Call-to-action (CTA) buttons - changing the text on the button, varying the sizes, colors, and page placement.

Marketers can facilitate personalization on any web device by creating a web experiment.

You can apply the following optional features when you create a web experiment:

  • Personalization - add personalization by applying an out-of-the-box web template, which eliminates the need for code, or start from scratch.

  • Page targeting - set the area of your website where you want the experiment to run. You can also add JavaScript to determine the context in which the experiment should execute.

  • Audience - you can configure the experiment to only run for a certain segment of guests. If you want to apply additional conditions for the experiment to run, you can use JavaScript.

  • Decisioning - associate a decision model with your experiment to recommend dynamic offers or content.

  • Previewing a web experiment - preview the experiment in your organization's website.

  • Test API calls and decisioning - compose an API response and test it.

  • Goals - define the goals for measuring the success of your web experiment in analytics.

  • Hypothesis - add a hypothesis to see if the intended outcome matches your results.

After you create a web experiment, you can set it to live, then monitor and analyze results.