pantheon.io

When JavaScript takes a long time to execute it can slow down website performance.

Duplicate JavaScript modules can slow down a website by requiring unnecessary downloads.

Inefficient animated content formats such as animated GIFs should be replaced by video.

This measures how long it takes a browser to render the first piece of content when someone visits a website.

This approximates when the main content of the page is visible to users.

Lazily loading above the fold images can negatively affect performance.

Avoid serving legacy JavaScript to modern browsers, only serving it to older browsers that require it.

When the main thread is busy the website may not respond quickly to user interactions.

This measures the time from when a user first interacts with a website to the time when the browser is actually able to respond to that interaction.

External scripts dynamically injected via document.write() can delay page load by tens of seconds.

The image used by the largest contentful paint element should be preloaded to maximize performance.

Avoid multiple page redirects as they introduce additional delays before the page can be loaded.

Delivering critical JavaScript and CSS inline and deferring all non-critical JavaScript and CSS can help prevent render blocking.

This indicates how quickly the contents of a page are visibly populated.

This measures the total amount of time that a page is blocked from responding to user input, such as mouse clicks, screen taps, or keyboard presses.

Excessively large page sizes can lead to long load times for visitors.

Minifying CSS files can reduce payload sizes.

Optimized images load faster and consume less network resources.

Incorrect character encoding can result in garbled text and is properly declared in the charset attribute or the Content-Type HTTP response header.

This measures unexpected page movements which visitors find annoying and can intefere with website functionality.

Specifying a doctype prevents the browser from switching to quirks mode, which can cause a website to render in unexpected ways.

Ensure text is user-visible and avoid flashes of invisible text while webfonts are loading.

Users are mistrustful or confused by websites that request their location without context.

Images should be responsive and proportional to the display size and the pixel ratio to maximize image clarity.

Unload event listeners do not fire reliably and listening for them can intefere with website functionality and browser optimizations.

Users are mistrustful of or confused by sites that request to send notifications without context.

Touch and wheel event listeners should be marked as passive to improve scroll performance.

Correctly setting the viewport optimizes websites for mobile screen sizes and also prevents delays to user input.

Inspector issues indicate unresolved problems which can include security issues.

All websites should be protected with HTTPS including ones that don't handle sensitive data.

Preventing password fields from being pasted into undermines good security policy.

Screen reader users rely on frame titles to describe the contents of frames.

If the website doesn't specify a language correctly screen readers might not be able to read the text correctly.

Specifying a valid land attribute helps screen readers announce text correctly.

Labels ensure that forms are announced properly by assistive technologies like screen readers.

Lists must only contain list elements so that screen readers can correctly announce lists.

List items must be contained within a parent list for screen readers to properly announce lists.

Disabling zooming and scaling is problematic for users with low vision who rely on screen magnification to properly see the contents of a website.

A tabindex value greater than 0 implies an explicit navigation ordering which often creates frustrating experiences for users who rely on assistive technologies.

Valid canonical links tell search engines which version of a page to crawl and display to users in search results.

Search engines use correctly formatted links to crawl websites.

Search engine users rely on the document title to determine if a page is relevant to their search.

Many sites provide different versions of a page based on a user's language or region and hreflang links tell search engines the URLs for all the versions of a page so that they can display the correct version.

Pages with unsuccessful HTTP status codes may not be indexed by search engines properly.

Search engines are unable to include websites in search results if they don't have permission to crawl them.

Search engines can't index plugin content and many devices restrict plugins or don't support them.

A valid robots.txt is required for search engines to understand how to crawl or index a website.