- Optimizely Web Experimentation
Although Optimizely Web Experimentation allows you to add a small snippet of code to your site to start experimenting immediately, you should follow this checklist to implement Optimizely Web Experimentation for long-term success, especially if this is your first time.
Use the checklist to perform an audit of your site's technical framework and complete your setup. You can complete most tasks once, unless major changes to your site or business model alter the technical specifications of your site. You can then build and run experiments and campaigns.
Document the technical specifications and initial setup in an implementation document and link it to your testing charter. You should update your technical audit document whenever your Optimizely Web Experimentation setup changes to help your team keep track of the types of experiments you can run in each project and troubleshoot.
Use this advanced implementation checklist to explore deeper functionality in Optimizely Web Experimentation.
You can also view a whitepaper on optimizing performance for client-side experimentation.
Site requirements checklist
You may need to change the activation mode on pages if you use Angular, Backbone, Ember, or Knockout. Pages are URL templates that tell Optimizely Web Experimentation where and how to activate experiments and Personalization campaigns.
When you set up pages in Optimizely Web Experimentation, use conditional or manual activation to ensure that experiments activate appropriately. See resources on how Optimizely Web Experimentation works with SPAs and using Optimizely Web Experimentation with UI libraries like React and Vue for more information.
Separate staging and production environments
If your team has separate staging and production environments, use a separate snippet and project for each environment. See how to set up Optimizely Web Experimentation for two environments.
If not, use the following steps to set up your production project:
Account settings checklist
Protect your Optimizely Web Experimentation account
Two-step verification, single sign-on, and password expiration settings help you keep your Optimizely Web Experimentation account secure.
Anonymize your visitors' IP addresses
In some countries, you may need to remove the last block of an IP address to protect the identity of your visitors. See enable IP anonymization for more information.
Project settings checklist
Sites with multiple origins: such as different subdomains or security protocols An “origin” is a combination of a specific hostname, security protocol, and port on your site.
By default, Optimizely Web Experimentation tracks events on the same origin. To track visitor behaviors across different origins, make sure you enable cross-origin targeting. If this is not set up, Optimizely Web Experimentation cannot track visitors across origins, such as subdomains and security protocols.
If your experiment depends on origin data being synced, you must use waitForOriginSync when switching origins.
Integrate technology platforms with Optimizely Web Experimentation
You can view a list of best-of-breed platforms that Optimizely Web Experimentation integrates with.
Optimizely Web Experimentation's integrations allow you to maximize the information you are getting from different data sources and give you the ability to segment results for business insights.
Manage multiple analytics integrations on the same page and target experiments on changes that happen on your page without reloading
Optimizely Web Experimentation allows you to customize your settings to increase privacy, help with easy QA and debugging, and ensure compliance with legal requirements.
Set up privacy settings for your account.
- On which URLs will you run experiments and campaigns on your site?
- On which URLs will you want to track visitor actions?
- How will you measure whether an experiment or campaign is successful?
See prepare your Optimizely Web Experimentation project to learn how to set up four major components for your site: pages, events, tags, and audiences.
You can add more later, but creating them now helps your organization run experiments with greater velocity.
Granular results for visitor segments
You may want to see more granular results for logged-in visitors or a list of customers in your database. The Results page gives you an aggregate of the change in behavior across all visitors. Because visitors are not all the same, you can segment results by characteristics that are important to your business with custom attributes.
Visitor actions not captured by clicks or pageviews
Some actions are not captured by clicks or pageviews, such as pop-up modal views, form submissions, or drop-offs in an order form. Custom events help you track actions beyond clicks and pageviews.
Talk to your product manager and analytics team to find key events. What behaviors determine success on each page? Revisit these on regular basis to consider parts of your site that have changed.
Custom visual elements like banners, carousels, and lightboxes
Extensions are reusable templates that help you make changes in the Visual Editor. A developer builds it once, then you can add and edit it without writing additional code.
Available on select Optimizely Web Experimentation packages.
You can set up revenue tracking to track revenue in Optimizely Web Experimentation.
Experiment with your payment or checkout page
Optimizely Web Experimentation is PCI-compliant, so you can safely experiment across your full checkout funnel.