About PageSpeed Insights

PageSpeed Insights reports on the real-world performance of a page or an origin for mobile and desktop devices and provides suggestions on how that page may be improved when analyzing a single page.

Speed Score

PageSpeed Insights incorporates data from the Chrome User Experience Report (CrUX) to display real-world performance data about a page or an origin. PSI reports two metrics: First Contentful Paint (FCP) and DOMContentLoaded (DCL).

The median value for each metric (FCP or DCL) is compared to all the pages monitored by the CrUX report. Each metric is assigned a category, depending on where it falls in the distribution:

  • Fast: The median value of the metric is in the fastest third of all page loads.
  • Slow: The median value of the metric is in the slowest third of all page loads.
  • Average: The median value of the metric is in the middle third of all page loads.

An overall Speed score is calculated by looking at the categories for each metric:

  • Fast: If every metric of a page/origin is Fast.
  • Slow: If any metric of a page/origin is Slow.
  • Average: All other cases.

Page Load Distribution

The Page Load Distribution section presents the distribution FCP and DCL metrics in the CrUX dataset.

  • Origin: Aggregated speed data of all pages served from this origin.
  • Page: The single-page's speed data.

The distribution for each metric is split into three categories: Fast (fastest third), Average (middle third), and Slow (bottom third). The distribution is updated on a daily basis.

Optimization Score

PageSpeed Insights only reports optimization information when analyzing a single page.

PSI evaluates how well a page follows common performance best practices and computes a score from 0-100 that estimates its performance headroom. It evaluates if a page can improve its performance in two areas:

  • time to above-the-fold load: Elapsed time from the moment a user requests a new page to the moment the above-the-fold content is rendered by the browser.
  • time to full page load: Elapsed time from the moment a user requests a new page to the moment the page is fully rendered by the browser.

The score is categorized as being Good, Medium, or Low. The calculation assumes that a developer wants to keep the same appearance and functionality of the page.

  • Good: The page applies most performance best practices and there is little headroom for further optimization. The page scores 80 or above.
  • Medium: The page is missing some common performance optimizations and there is medium headroom for optimization. The page scores between 60 and 79.
  • Low: The page is not optimized and there is fairly large headroom for optimization. The page scores between 0 and 59.

The optimization score estimates the performance headroom of a page. A page with slow speed might have high optimization score because it has a low ratio of render-blocking resources relative to the total resources used by the page. On the other hand, a fast page might have a low optimization score because it has a high number of render-blocking resources relative to its total resources.

Page Stats

This section describes the round trips required to load the page’s render-blocking resources, the total bytes used by the page, and how it compares to the median number of round trips and bytes used in the dataset. It can indicate if the page might be faster by modifying the appearance and functionality of the page.

Optimization Suggestions

These are a list of best practices that could be applied to this page. If the page is fast, these suggestions are hidden by default, as the page is already in the top third of all pages in the data set, and further improvement is likely unnecessary.

Since the performance of a network connection varies considerably, the optimization addresses the network-independent aspects of page performance: the server configuration, the HTML structure of a page, and its use of external resources such as images, JavaScript, and CSS. Implementing the suggestions should improve the relative performance of the page. However, the absolute performance of the page will still be dependent upon a user’s network connection.

The speed performance is affected by various factors, but two of them are most important:

Questions?

If you have questions about PageSpeed Insights, please refer to our FAQ or post a message to our discussion group.