Skip to main content

Get started

Welcome to the Lyftio documentation​

This guide will help you set up your first experiment from start to finish. By the end, you’ll have a working experiment running on your website with a targeted audience, defined goals, and live variations.

What you’ll do in this guide

  • Create your project
    Before installing the Lyftio script on your website, you must first create a project inside Lyftio. Each project has its own unique script snippet, which you’ll find in the project settings once the project is created.

  • Install the script
    Add the project-specific Lyftio script to your website so that experiments can be delivered, targeted, and tracked correctly.

  • Project setup
    When creating a new project, you’ll configure how Lyftio handles visitor data and consent. These settings determine how experiments are delivered and how visitor interactions are tracked.

    • Select the level of consent enforcement that matches your organization’s policies
    • Choose where Lyftio should store visitor data. This determines where each visitor’s unique ID, assigned experiments, and other experiment-related information will be saved in their browser while interacting with your site.

  • Create your first experiment
    Give it a name, an optional description and set the page you want to open the editor on.

  • Set up your audience
    Decide who should see the experiment and for which page(s) the variation changes should happen. This is done by setting conditions that the visitor needs to fulfill to be part of the experiment.

  • Define your goals
    Set the KPIs that track what matters to your business, such as clicks, sign-ups, or purchases. These are the metrics used to measure experiment success. Goals also use conditions that the visitor must meet for the goal to register as a conversion.

  • Build your variations
    Use the Visual Editor to make changes without touching code. Add new text, swap images, or completely redesign elements to create meaningful alternatives for your visitors.

  • Verify your experiment with Preview Make sure your new variations look and work as expected before launching the experiment for your visitors. Test that all you goals trigger as they should for all devices that the experiment will run for.

  • Launch the experiment
    Once everything looks good, start the experiment to test your hypotheses.

  • Analyse the result
    See how the variation(s) perfomed against your original.

Whether you’re running your very first A/B test or scaling up to multiple experiments, this guide will give you the foundation you need to make confident, data-driven decisions.