Load Third-Party JavaScript

Arthur Evans

If you've optimized your code, but your site still loads too slowly, it's probably the fault of third-party scripts.

Third-party scripts provide a variety of useful features that make the web more dynamic, interactive, and interconnected. Some of them might even be crucial to your website's function or revenue stream. But using them is risky:

  • They can slow down your site's performance.
  • They can cause privacy or security issues.
  • They can be unpredictable and their behavior can have unintended consequences.

Ideally, you'll want to ensure third-party scripts aren't impacting your site's critical rendering path. In this guide, we'll walk through how to find and fix issues related to loading third-party JavaScript and minimize the risks to your users.

What are third-party scripts?

Third-party JavaScript often refers to scripts that can be embedded into any site directly from a third-party vendor. Examples include:

  • Social sharing buttons (Facebook, X, LinkedIn, Mastodon)

  • Video player embeds (YouTube, Vimeo)

  • Advertising iframes

  • Analytics & metrics scripts

  • A/B testing scripts for experiments

  • Helper libraries, such as date formatting, animation, or functional libraries

example of a youtube video embed
An example YouTube video embed that can be added to a page using the following code.
<iframe
  width="560"
  height="315"
  src="https://www.youtube.com/embed/mo8thg5XGV0"
  frameborder="0"
  allow="autoplay; encrypted-media"
  allowfullscreen
>
</iframe>

Unfortunately, embedding third-party scripts means we often rely on them to run quickly and not slow our pages down. Third-party scripts are a common cause of performance slowdowns caused by resources outside the site owner's control, including the following issues:

  • Firing too many network requests to multiple servers. The more requests a site has to make, the longer it can take to load.

  • Sending too much JavaScript that keeps the main thread busy. Too much JavaScript can block DOM construction, which delays page rendering. CPU-intensive script parsing and execution can delay user interaction and cause battery drain.

  • Sending large, unoptimized image files or videos can consume data and cost users money.

  • Security issues that can act as a single-point of failure (SPOF) if your page loads a script without caution.

  • Insufficient HTTP caching, forcing the browser to send more network requests to fetch resources.

  • Lack of sufficient server compression causes resources to load slowly.

  • Blocking content display until they complete processing. This can also be true for async A/B testing scripts.

  • Use of legacy APIs known to harm the user experience, such as document.write().

  • Excessive DOM elements or expensive CSS selectors.

  • Including multiple third-party embeds can lead to multiple frameworks and libraries being pulled in several times, wasting resources and making existing performance issues worse.

  • Third-party scripts often use embed techniques that can block window.onload if their servers respond slowly, even if the embed is using async or defer.

Your ability to fix issues with third-party scripts can depend on your site and your ability to configure how you load third-party code. Fortunately, a number of solutions and tools exist to find and fix issues with third-party resources.

How do you identify third-party script on a page?

Identifying third-party scripts on your site and determining their performance impact is the first step toward optimizing them. We recommend using free web speed test tools, including Chrome DevTools, PageSpeed Insights, and WebPageTest, to identify costly scripts. These tools display rich diagnostic information that can tell you how many third-party scripts your site uses, and which take the most time to execute.

WebPageTest's waterfall view can highlight the impact of heavy third-party script use. The following image from Tags Gone Wild shows an example diagram of the network requests required to load the main content for a site, as opposed to the tracking and marketing scripts.

waterfall view from webpagetest showing an
actual website vs the amount of time spent loading tracking scripts
The scripts on this page take longer to load than the page itself.

WebPageTest's domain breakdown can also be useful for visualizing how much content comes from third-party origins. It breaks this down by both total bytes and the number of requests:

content breakdown by domain (first view).
Shows the percentage of requests and bytes for each third-party
The domain breakdown charts show how much of a page's content comes from third parties.

How do I measure the impact of third-party script on my page?

When you see a script causing issues, find out what the script does and determine whether your site needs it to function. If necessary, run an A/B test to balance its perceived value versus its impact on key user engagement or performance metrics.

Lighthouse Boot-up Time Audit

The Lighthouse JavaScript boot-up time audit highlights scripts that have a costly script parse, compile or evaluation time. This can help you identify CPU-intensive third-party scripts.

Lighthouse showing support for script
evaluating and parsing
The boot-up time audit shows which scripts take longest to load.

Lighthouse Network Payloads Audit

The Lighthouse Network Payloads Audit identifies network requests, including third-party network requests that slow down page load time and make users spend more than they expect on mobile data.

Lighthouse showing support for large network
payloads
The Network Payloads Audit shows which network requests take the most time and fetch the most data.

Chrome DevTools Network Request Blocking

Chrome DevTools lets you see how your page behaves when a specified script, style sheet, or other resource isn't available. This is done with network request blocking, a feature that can help measure the impact of removing individual third-party resources from your page.

To enable request blocking, right-click any request in the Network panel and select Block Request URL. A request blocking tab then displays in the DevTools drawer, letting you manage which requests have been blocked.

Block request URLs from the DevTools network
panel
Block individual network requests to see how your page behaves without them.

Chrome DevTools Performance Panel

The Performance panel in Chrome DevTools helps identify issues with your page's web performance.

  1. Click Record.
  2. Load your page. DevTools shows a waterfall diagram representing how your site spends its loading time.
  3. Navigate to Bottom-up at the bottom of the Performance panel.
  4. Click Group by product and sort your page's third-party scripts by load time.
DevTools Performance panel showing the
Bottom-up view grouped by (third-party) products
Third-party scripts sorted by product, starting with the longest load time.

To learn more about using Chrome DevTools to analyze page load performance, see Get started with analyzing runtime performance.

The following is our recommended workflow for measuring the impact of third-party scripts:

  1. Use the Network panel to measure how long it takes to load your page.
    • To emulate real-world conditions, we recommend turning on network throttling and CPU throttling. Your users are unlikely to have the fast network connections and desktop hardware that reduce the impact of expensive scripts under lab conditions.
  2. Block the URLs or domains responsible for third-party scripts you believe are an issue (see Chrome DevTools Performance Panel for guidance on identifying costly scripts).
  3. Reload the page and measure the load time again.
    • For more accurate data, you might want to measure your load time at least three times. This accounts for some third-party scripts fetching different resources on each page load. To help with this, the DevTools Performance Panel supports multiple recordings.

Measure third-party scripts' impact with WebPageTest

WebPageTest supports blocking individual requests from loading to measure their impact in Advanced Settings > Block. Use this feature to specify a list of domains to block, such as advertising domains.

WebPageTest advanced settings < Block.
Displays a text area for specifying domains to block.
List the domains you want to block in this panel.

We recommend the following workflow to use this feature:

  1. Test a page without blocking third parties.
  2. Repeat the test with some third parties blocked.
  3. Select the two results from your Test History.
  4. Click Compare.
WebPageTest displaying the compare option
allowing you to compare two reports
Select load test results to compare.

The following image shows WebPageTest's filmstrip feature comparing the load sequences for pages with and without active third-party resources. We recommend checking this for tests of individual third-party origins, to determine which domains impact your page's performance most.

WebPageTest filmstrip displaying the impact
of loading a site with and without third parties disabled
The impact of blocking third-party resources, from Using WebPageTest To Measure The Impact Of Third-Party Tags by Andy Davies.

WebPageTest also supports two commands operating at the DNS level for blocking domains:

WebPageTest also has a single point of failure (SPOF) tab that lets you simulate a timeout or complete failure to load a resource. Unlike domain blocking, SPOF slowly times out, which can make it useful for testing how your pages behave when third-party services are under heavy load or temporarily unavailable.

WebPageTest advanced settings > SPOF > hosts
to fail
Use the SPOF testing feature to simulate the failure of specified domains.

Detect expensive iframes using Long Tasks

When scripts in third-party iframes take a long time to run, they can block the main thread and delay other tasks. These long tasks can cause event handlers to work slowly or frames to drop, making the user experience worse.

To detect long tasks for Real User Monitoring (RUM), use the JavaScript PerformanceObserver API to observe longtask entries. These entries contain an attribution property that you can use to determine which frame context caused the long task.

The following code logs longtask entries to the console, including one for an "expensive" iframe:

<script>
  const observer = new PerformanceObserver((list) => {
    for (const entry of list.getEntries()) {
      // Attribution entry including "containerSrc":"https://example.com"
      console.log(JSON.stringify(entry.attribution));
    }
  });

  observer.observe({entryTypes: ['longtask']});
</script>

<!-- Imagine this is an iframe with expensive long tasks -->
<iframe src="https://example.com"></iframe>

To learn more about monitoring Long Tasks, see User-centric Performance Metrics.

How do you load third-party script efficiently?

If a third-party script is slowing down your page load, you have several options to improve performance:

  • Load the script using the async or defer attribute to avoid blocking document parsing.
  • If the third-party server is slow, consider self-hosting the script.
  • If the script doesn't add clear value to your site, remove it.
  • Use Resource Hints like <link rel=preconnect> or <link rel=dns-prefetch> to perform a DNS lookup for domains hosting third-party scripts.

Use async or defer

JavaScript execution is parser-blocking. When the browser encounters a script, it must pause DOM construction, pass the script to the JavaScript engine, and allow the script to execute before continuing with DOM construction.

The async and defer attributes change this behavior as follows:

  • async causes the browser to download the script asynchronously while it continues to parse the HTML document. When the script finishes downloading, parsing is blocked while the script executes.

  • defer causes the browser to download the script asynchronously while it continues to parse the HTML document, then waits to run the script until document parsing is complete.

Always use async or defer for third-party scripts, unless the script is necessary for the critical rendering path. Use async if it's important for the script to run earlier in the loading process, such as for some analytics scripts. Use defer for less critical resources, such as videos that render lower on the page than the user will initially see.

If performance is your primary concern, we recommend waiting to add asynchronous scripts until after your page's critical content loads. We don't recommend using async for essential libraries such as jQuery.

Some scripts must be loaded without async or defer, especially those which are crucial parts of your site. These include UI libraries or content delivery network (CDN) frameworks that your site can't function without.

Other scripts just don't work if they're loaded asynchronously. Check the documentation for any scripts you're using, and replace any scripts that can't be loaded asynchronously with alternatives that can. Be aware that some third parties recommend running their scripts synchronously, even if they would work equally well asynchronously.

Remember that async doesn't fix everything. If your page includes a large number of scripts, such as tracking scripts for advertising purposes, loading them asynchronously won't prevent them from slowing down page load.

Use Resource Hints to reduce connection setup time

Establishing connections to third-party origins can take a long time, especially on slow networks, because network requests have multiple complex components, including DNS lookups and redirects. You can use Resource Hints like to perform DNS lookups for domains hosting third-party scripts early in the page load process, so that the rest of the network request can proceed more quickly later:

<link rel="dns-prefetch" href="http://example.com" />

If the third-party domain you're connecting to uses HTTPS, you can also use , which both performs DNS lookups and resolves TCP round-trips and handle TLS negotiations. These other steps can be very slow because they involve verifying SSL certificates, so preconnecting can greatly decrease load time.

<link rel="preconnect" href="https://cdn.example.com" />

"Sandbox" scripts with an iframe

If you load a third-party script directly into an iframe, it doesn't block execution of the main page. AMP uses this approach to keep JavaScript out of the critical path. Note that this approach still blocks the onload event, so try not to attach critical features to onload.

Chrome also supports Permissions Policy (formerly Feature Policy), a set of policies that allow a developer to selectively disable access to certain browser features. You can use this to prevent third-party content from introducing unwanted behaviors to a site.

Self-host third-party scripts

If you want more control over how a critical script loads, for example to reduce DNS time or improve HTTP caching headers, you might be able to host it yourself.

However, self-hosting comes with issues of its own, especially when it comes to updating scripts. Self-hosted scripts won't get automatic updates for API changes or security fixes, which can lead to revenue losses or security issues until you can update your script manually.

As an alternative, you can cache third-party scripts using service workers for more control over how often scripts are fetched from the network. You can also use service workers to create loading strategies that throttle nonessential third-party requests until your page reaches a key user moment.

A/B Test smaller samples of users

A/B testing (or split-testing) is a technique for experimenting with two versions of a page to analyze user experience and behavior. It makes the page versions available to different samples of your website traffic, and determines from analytics which version provides a better conversion rate.

However, by design, A/B testing delays rendering to decide which experiment needs to be active. JavaScript is often used to check if any of your users belong to an A/B test experiment and then enable the correct variant. This process can worsen the experience even for users who aren't part of the experiment.

To speed up page rendering, we recommend sending your A/B testing scripts to a smaller sample of your user base, and run the code that decides which version of the page to display serverside.

Lazy load Third-Party Resources

Embedded third-party resources, such as ads and videos, can be major contributors to slow page speed when constructed poorly. Lazy loading can be used to load embedded resources only when necessary, for example, waiting to serve ads in the page footer until the user scrolls far enough to see them. You can also lazy load third-party content after the main page content loads but before a user might otherwise interact with the page.

An illustration showing assets that are
critical for the above the fold experience and those that are less critical and
can be lazily loaded in.
You can lazy load assets that the user won't see immediately when the page loads.

Be careful when lazy loading resources, because it often involves JavaScript code that can be impacted by flaky network connections.

DoubleClick have guidance on how to lazy load ads in their official documentation.

Efficient lazy loading with Intersection Observer

Historically, methods for detecting whether an element is visible in the viewport for lazy loading purposes have been error-prone and often slow down the browser. These inefficient methods often listen for scroll or resize events, then use DOM APIs like getBoundingClientRect() to calculate where elements are relative to the viewport.

IntersectionObserver is a browser API that lets page owners efficiently detect when an observed element enters or leaves the browser's viewport. LazySizes also has optional support for IntersectionObserver.

Lazy load analytics

If you defer loading of your analytics scripts for too long, you can miss valuable analytics data. Fortunately, there are strategies available for initializing analytics lazily while retaining early page-load data.

Phil Walton's blog post The Google Analytics Setup I Use on Every Site I Build covers one such strategy for Google Analytics.

Load third-party scripts safely

This section provides guidance for loading third-party scripts as safely as possible.

Avoid document.write()

Third-party scripts, especially for older services, sometimes use document.write() to inject and load scripts. This is a problem because document.write() behaves inconsistently, and its failures are difficult to debug.

The fix for document.write() issues is to not use it. In Chrome 53 and higher, Chrome DevTools logs warnings to the console for problematic use of document.write():

DevTools console warnings highlighting
violations for a third-party embed using document.write()
Chrome DevTools flags document.write() usage.

If you receive this error, you can check your site for document.write() usage by looking for HTTP headers sent to your browser. Lighthouse can also highlight any third-party scripts still using document.write().

Lighthouse Best Practices audit flagging use
of document.write()
A Lighthouse report showing which scripts use document.write().

Use Tag Managers carefully

A tag is a code snippet that lets digital marketing teams collect data, set cookies, or integrate third-party content like social media widgets into a site. These tags add network requests, JavaScript dependencies, and other resources to your page that can impact its performance, and minimizing that impact for your users gets more difficult as more tags are added.

To keep page loading fast, we recommend using a tag manager such as Google Tag Manager (GTM). GTM lets you deploy tags asynchronously so they don't block each other from loading, reduces the number of network calls a browser needs to execute tags, and collects tag data in its Data Layer UI.

Risks of using tag managers

Although tag managers are designed to streamline page loading, using them carelessly can slow it down instead in the following ways:

  • Excessive numbers of tags and auto-event listeners in your tag manager causes the browser to make more network requests than it would otherwise need to, and reduces your code's ability to respond to events quickly.
  • Anyone with credentials and access can add JavaScript to your tag manager. Not only can this increase the number of costly network requests needed to load your page, but can also present security risks and other performance issues from unnecessary scripts. To reduce these risks, we recommend limiting access to your tag manager.

Avoid scripts that pollute the global scope

Third-party scripts can behave in all kinds of ways that break your page unexpectedly:

  • Scripts loading JavaScript dependencies can pollute the global scope with code that interacts badly with your code.
  • Unexpected updates can cause breaking changes.
  • Third-party code can be modified in transit to behave differently between testing and deployment of your page.

We recommend conducting regular audits of the third-party scripts you load to check for bad actors. You can also implement self-testing, subresource integrity, and secure transmission of third-party code to keep your page safe.

Mitigation strategies

Here are some large-scale strategies for minimizing the impact of third-party scripts on your site's performance and security:

  • HTTPS: Sites that use HTTPS must not depend on third parties that use HTTP. For more information, refer to Mixed Content.

  • Sandboxing: Consider running third-party scripts in iframes with the sandbox attribute to restrict the actions available to the scripts.

  • Content Security Policy (CSP): You can use HTTP headers in your server's response to define trusted script behaviors for your site, and detect and mitigate the effects of some attacks, such as Cross Site Scripting (XSS).

The following is an example of how to use CSP's script-src directive to specify a page's allowed JavaScript sources:

// Given this CSP header Content-Security-Policy: script-src
https://example.com/ // The following third-party script will not be loaded or
executed

<script src="https://not-example.com/js/library.js"></script>

Further reading

To learn more about optimizing third-party Javascript, we recommend you read the following:

With thanks to Kenji Baheux, Jeremy Wagner, Pat Meenan, Philip Walton, Jeff Posnick and Cheney Tsai for their reviews.