Skip to main content

Architecture Overview

Optimize Deprecation

Uniform Optimize has been deprecated and replaced by Context, a more powerful and flexible personalization solution. Optimize is not being discontinued at this time, but it will not receive updates or new features.

We do not recommend starting a new project with Optimize. If you have an existing project that uses Optimize, you can upgrade your project to Context at no cost using our upgrade guide. If you have any issues with this process you can contact our team.

Jamstack Native​

Uniform Optimize is designed for Jamstack sites: personalization data, the intent manifest,  is baked in at build time. The execution of personalization is ridiculously fast compared to traditional personalization solutions. Personalization loading in only a few milliseconds? Yes, please.

tip

Uniform is flexible in its execution model; it can run at the edge, purely client-side, or with server-side-rendering.

Privacy First​

Uniform Optimize stores all personalization data and history on the visitor's browser by default. It gives the visitor complete control over the data you have collected about their personalization profile.

Want to store personalization data for your analysis? Our extensible plugin architecture lets you put data wherever you want.

Data Sources​

Uniform Optimize uses two primary sources of data:

  • The intent manifest - defines an array of known intents and their signals. It is usually provided at build time by a Uniform Optimize request and baked into your JS bundle.
  • The visitor intent scoring - defines the score of the current visitor in each intent. These scores are based on triggering signals that modify their intent score. The intent score is stored most commonly in cookies. The most common types of signals are:
    • Request signals - cookies, query strings, page route, visited, etc. These are triggered based on the browser's request.
    • Behavior signals - triggered based on content with intent tags shown to the visitor. For example, being shown a call to action with a 'developer' tag will result in some developer score being added.
    • Programmatic signals - Event signals are matched on an event stream such as Google Analytics events (i.e., 'label contains foo'). Developers connect this event stream.

These data sources are used to perform visitor classification. The intent manifest provides the rules by which we calculate intent scoring. Scoring, in turn, influences how we select a personalized variation to show visitors.