README.md

Third Party Web

This document is a summary of which third party scripts are most responsible for excessive JavaScript execution on the web today.

Table of Contents

1. Goals
2. Methodology
3. Data
   1. Summary
   2. How to Interpret
   3. Third Parties by Category
      1. Ads
      2. Analytics
      3. Social
      4. Video
      5. Developer Utilities
      6. Hosting Platforms
      7. Marketing
      8. Customer Success
      9. Content & Publishing
      10. Libraries
      11. Mixed / Other
   4. Third Parties by Total Impact
4. Future Work
5. FAQs
6. Contributing

Goals

1. Quantify the impact of third party scripts on the web.
2. Identify the third party scripts on the web that have the greatest performance cost.
3. Give developers the information they need to make informed decisions about which third parties to include on their sites.
4. Incentivize responsible third party script behavior.

Methodology

HTTP Archive is an inititiave that tracks how the web is built. Twice a month, ~4 million sites are crawled with Lighthouse on mobile. Lighthouse breaks down the total script execution time of each page and attributes the execution to a URL. Using BigQuery, this project aggregates the script execution to the origin-level and assigns each origin to the responsible entity.

Data

Summary

Across top ~1 million sites, ~800 origins account for ~65% of all script execution time with the top 100 entities already accounting for ~59%. Third party script execution is the majority chunk of the web today, and it's important to make informed choices.

How to Interpret

Each entity has a number of data points available.

1. Usage (Total Number of Occurrences) - how many scripts from their origins were included on pages
2. Total Impact (Total Execution Time) - how many seconds were spent executing their scripts across the web
3. Average Impact (Average Execution Time) - on average, how many milliseconds were spent executing each script
4. Category - what type of script is this

Third Parties by Category

This section breaks down third parties by category. The third parties in each category are ranked from first to last based on the average impact of their scripts. Perhaps the most important comparisons lie here. You always need to pick an analytics provider, but at least you can pick the most well-behaved analytics provider.

Overall Breakdown

Unsurprisingly, ads account for the largest identifiable chunk of third party script execution. Other balloons as a category primarily due to Google Tag Manager which is used to deliver scripts in multiple categories. Google Tag Manager script execution alone is responsible for more than half of the "Mixed / Other" category.

Ads

These scripts are part of advertising networks, either serving or measuring.

Rank Name Usage Average Impact 1 Scorecard Research 4,740 90 ms 2 Criteo 63,058 137 ms 3 Taboola 23,818 183 ms 4 AppNexus 16,942 235 ms 5 Yahoo Ads 8,495 247 ms 6 Pubmatic 3,033 255 ms 7 Market GID 3,831 274 ms 8 MGID 10,472 277 ms 9 Integral Ads 23,942 296 ms 10 Google/Doubleclick Ads 1,412,404 330 ms 11 Sizmek 3,831 333 ms 12 Yandex Ads 28,882 386 ms 13 DoubleVerify 4,041 503 ms 14 Moat 23,170 616 ms 15 OpenX 12,505 821 ms 16 MediaVine 9,205 839 ms 17 Media Math 3,498 905 ms 18 33 Across 20,756 1170 ms 19 Popads 6,545 1245 ms 20 WordAds 30,961 2543 ms

Analytics

These scripts measure or track users and their actions. There's a wide range in impact here depending on what's being tracked.

Rank Name Usage Average Impact 1 Alexa 1,721 56 ms 2 Baidu Analytics 8,018 79 ms 3 Mixpanel 7,258 79 ms 4 Google Analytics 1,533,217 79 ms 5 Hotjar 108,933 89 ms 6 Crazy Egg 2,214 91 ms 7 Adobe Analytics 34,436 192 ms 8 Tealium 15,933 214 ms 9 Segment 6,581 228 ms 10 Optimizely 13,853 247 ms 11 Salesforce 40,451 291 ms 12 Histats 15,770 339 ms 13 Yandex Metrica 217,229 377 ms 14 Lucky Orange 6,037 870 ms

Social

These scripts enable social features.

Rank Name Usage Average Impact 1 VK 7,578 61 ms 2 Pinterest 11,122 71 ms 3 Facebook 1,212,567 120 ms 4 Yandex Share 30,677 128 ms 5 Twitter 295,308 154 ms 6 LinkedIn 10,663 156 ms 7 ShareThis 39,884 216 ms 8 AddThis 179,424 254 ms 9 Tumblr 46,800 330 ms 10 PIXNET 53,889 473 ms 11 Disqus 723 580 ms

Video

These scripts enable video player and streaming functionality.

Rank Name Usage Average Impact 1 YouTube 23,184 104 ms 2 Wistia 21,319 254 ms 3 Brightcove 5,133 469 ms

Developer Utilities

These scripts are developer utilities (API clients, site monitoring, fraud detection, etc).

Rank Name Usage Average Impact 1 New Relic 3,613 52 ms 2 Stripe 3,775 66 ms 3 OneSignal 41,178 88 ms 4 Google APIs/SDK 977,712 115 ms 5 App Dynamics 2,248 117 ms 6 Cloudflare 3,151 185 ms 7 PayPal 6,376 241 ms 8 Yandex APIs 57,425 368 ms 9 Distil Networks 11,489 409 ms 10 Sentry 15,272 729 ms

Hosting Platforms

These scripts are from web hosting platforms (WordPress, Wix, Squarespace, etc). Note that in this category, this can sometimes be the entirety of script on the page, and so the "impact" rank might be misleading. In the case of WordPress, this just indicates the libraries hosted and served by WordPress not all sites using self-hosted WordPress.

Rank Name Usage Average Impact 1 WordPress 135,176 113 ms 2 Shopify 227,933 163 ms 3 Squarespace 86,605 412 ms 4 Hatena Blog 53,675 516 ms 5 Wix 158,466 1153 ms

Marketing

These scripts are from marketing tools that add popups/newsletters/etc.

Rank Name Usage Average Impact 1 Hubspot 15,991 89 ms 2 OptinMonster 1,483 127 ms 3 Beeketing 62,659 149 ms 4 Drift 3,929 156 ms 5 Mailchimp 21,946 161 ms 6 Sumo 39,747 390 ms 7 Albacross 1,407 769 ms

Customer Success

These scripts are from customer support/marketing providers that offer chat and contact solutions. These scripts are generally heavier in weight.

Rank Name Usage Average Impact 1 LiveChat 23,881 93 ms 2 Freshdesk 973 154 ms 3 Help Scout 665 198 ms 4 Olark 14,917 309 ms 5 Tawk.to 40,228 386 ms 6 ZenDesk 36,411 476 ms 7 Intercom 15,511 554 ms 8 Zopim 55,964 688 ms

Content & Publishing

These scripts are from content providers or publishing-specific affiliate tracking.

Rank Name Usage Average Impact 1 AMP 60,944 196 ms 2 Vox Media 702 558 ms 3 Hotmart 1,008 814 ms

Libraries

These are mostly open source libraries (e.g. jQuery) served over different public CDNs. This category is unique in that the origin may have no responsibility for the performance of what's being served. Note that rank here does not imply one CDN is better than the other. It simply indicates that the libraries being served from that origin are lighter/heavier than the ones served by another..

Rank Name Usage Average Impact 1 Bootstrap CDN 2,860 44 ms 2 FontAwesome CDN 17,002 102 ms 3 Yandex CDN 2,499 115 ms 4 jQuery CDN 170,001 154 ms 5 Cloudflare CDN 119,800 176 ms 6 Google CDN 811,231 178 ms 7 JSDelivr CDN 27,070 257 ms 8 CreateJS CDN 1,988 3188 ms

Mixed / Other

These are miscellaneous scripts delivered via a shared origin with no precise category or attribution. Help us out by identifying more origins!

Rank Name Usage Average Impact 1 Amazon S3 35,291 152 ms 2 All Other 3rd Parties 1,649,095 198 ms 3 Google Tag Manager 1,093,167 386 ms 4 Parking Crew 4,021 428 ms

Third Parties by Total Impact

This section highlights the entities responsible for the most script execution across the web. This helps inform which improvements would have the largest total impact.

Future Work

1. Introduce URL-level data for more fine-grained analysis, i.e. which libraries from Cloudflare/Google CDNs are most expensive.
2. Expand the scope, i.e. include more third parties and have greater entity/category coverage.

FAQs

I don't see entity X in the list. What's up with that?

This can be for one of several reasons:

1. The entity does not have at least 100 references to their origin in the dataset.
2. The entity's origins have not yet been identified. See How can I contribute?

How is the "Average Impact" determined?

The HTTP Archive dataset includes Lighthouse reports for each URL on mobile. Lighthouse has an audit called "bootup-time" that summarizes the amount of time that each script spent on the main thread. The "Average Impact" for an entity is the total execution time of scripts whose domain matches one of the entity's domains divided by the total number of occurences of those scripts.

Average Impact = Total Execution Time / Total Occurences

How does Lighthouse determine the execution time of each script?

Lighthouse's bootup time audit attempts to attribute all toplevel main-thread tasks to a URL. A main thread task is attributed to the first script URL found in the stack. If you're interested in helping us improve this logic, see Contributing for details.

The data for entity X seems wrong. How can it be corrected?

Verify that the origins in data/entities.json are correct. Most issues will simply be the result of mislabelling of shared origins. If everything checks out, there is likely no further action and the data is valid. If you still believe there's errors, file an issue to discuss futher.

How can I contribute?

Only about 90% of the third party script execution has been assigned to an entity. We could use your help identifying the rest! See Contributing for details.

Contributing

Updating the Entities

The origin->entity mapping can be found in data/entities.json. Adding a new entity is as simple as adding a new array item with the following form.

{
    "name": "Facebook",
    "homepage": "https://www.facebook.com",
    "categories": ["social"],
    "origins": [
        "www.facebook.com",
        "connect.facebook.net",
        "staticxx.facebook.com",
        "static.xx.fbcdn.net",
        "m.facebook.com"
    ]
}

Updating Attribution Logic

The logic for attribution to individual script URLs can be found in the Lighthouse repo. File an issue over there to discuss further.

Updating the Data

The query used to compute the origin-level data is in sql/origin-execution-time-query.sql, running this against the latest Lighthouse HTTP Archive should give you a JSON export of the latest data that can be checked in at data/YYYY-MM-DD-origin-scripting.json.

Updating this README

This README is auto-generated from the template lib/template.md and the computed data. In order to update the charts, you'll need to make sure you have cairo installed locally in addition to yarn install.

# Install `cairo` and dependencies for node-canvas
brew install pkg-config cairo pango libpng jpeg giflib