Navigation AI is Uxify’s optimization layer that makes website navigations feel instant. It uses the Speculation Rules API together with our own AI engine, which learns from visitor behavior and your site’s historical data. The result is faster page transitions improving your site’s performance.
How Navigation AI works?
Navigation AI optimizes user navigations that happen after the initial page load. It uses a small, non-blocking script to ensure it won't slow down your site's core performance. It predicts the next page your visitor is most likely to open. It then preloads that page in the background so it’s ready by the time they click.
It does this by combining:
Real-time visitor behavior – analyzing how users move on your site.
Historical traffic data – learning from past navigation patterns.
Speculation Rules API – the browser feature that enables safe preloading.
This approach ensures that preload activity is targeted and efficient, giving the visitor a faster experience without unnecessary resource usage.
How to install Navigation AI?
Navigation AI is included automatically when you install the Uxify snippet.
No separate setup is required.
Once the snippet is added, Navigation AI can start analyzing visitor flows and optimizing navigations.
Dashboards in Navigation AI
Inside the Uxify app, the Navigation AI section gives you a full overview of its impact. You’ll see several dashboards:
Overview
The Overview dashboard shows how often Navigation AI is actively improving user journeys. Use the dropdown to switch between four metrics, depending on whether you want a session-level or navigation-level view, and whether to include only compatible traffic or your entire audience.
Optimized user sessions
What it shows?
The percentage of sessions on compatible browsers that Navigation AI successfully optimized. “Optimized user session” means the session was eligible based on browser support, non-excluded pages, and the presence of navigations that can be enhanced.
Why it matters?
A clean read on effectiveness when Navigation AI can actually run. It answers: When eligible, how often do we optimize?
Best for:
Measuring feature performance without browser noise
Testing changes to excludes, rules, or page templates
What influences it?
Browser support for speculative loading, your exclude rules, link patterns, and clarity of next steps on key pages.
Optimized navigations
What it shows?
The percentage of eligible navigations that Navigation AI optimized. Includes only compatible browsers and excludes sensitive flows (e.g., checkout, login, account).
Why it matters?
An event-level effectiveness view. It answers: Of the clicks we could enhance, how many did we enhance?
Best for:
Fine-tuning speculation rules and excludes
Spotting page types or paths where predictions don’t trigger
What influences it?
Prediction quality on specific flows, link markup, and intent signals, and exclusion settings.
Experience wins
Compares Core Web Vitals (CWV) for optimized vs. standard navigations, such as:
LCP (Largest Contentful Paint) – speed of loading
INP (Interaction to Next Paint) – speed of interaction
CLS (Cumulative Layout Shift) – visual stability
TTFB (Time to First Byte) – server response time
For example, you may see LCP reduced from ~2s to 0.3s on optimized navigations.
Engagement wins
Shows how performance improvements affect behavior:
Time to engage – how quickly users interact after a page loads.
Prevented loading exits – sessions saved from users leaving too early.
Pageviews per session – how faster journeys encourage deeper engagement.
Settings
Configure how Navigation AI runs:
Optimization mode – choose between Eco (efficiency) or Max (maximum preloading).
Excludes – remove specific URLs from optimization (useful for sensitive or dynamic pages).
Why it matters?
Navigation AI bridges the gap between performance engineering and user experience. By turning predictive intelligence into real speed gains, it improves Core Web Vitals, reduces exits, and drives higher engagement.