Skip to content
Audit Complete

Results for https://openai.com

Visit site
Netherlands Netherlands · Amsterdam Completed: Apr 16, 2026 17:18 UTC
Download JSON Download Markdown Report

Mobile 375 × 812

Screenshot of the audited page (Mobile 375×812)

Desktop 1440 × 900

Screenshot of the audited page (Desktop 1440×900)
C71

Site Health

Score: 71 / 100

Based on 8 categories, 0 sections

Decent speed, but optimizing further could improve engagement.

Good foundation, but a few gaps could be exploited.

Several issues make your site difficult for assistive technology users.

Solid SEO basics — a few optimizations could boost rankings.

Solid infrastructure — fast server responses across the board.

Several regulatory requirements are not yet met.

Missing metadata means poor previews on social media and search.

Heavier than average — reducing page weight saves energy and bandwidth.

How is this calculated?

The overall score is a weighted average of individual category scores. Categories with more impact on user experience and security carry more weight.

Performance 25%Security 25%Accessibility 15%SEO 10%Infrastructure 10%Compliance 8%Content 5%Sustainability 2%

Weights reflect general web best practices. Individual needs may differ.

How the composite score is calculated

How you compare

AngularJS sites average 73
At average
Better than 20% of AngularJS sites See full AngularJS benchmark →
Apple sites average 73
At average
Better than 18% of Apple sites See full Apple benchmark →

Top Priorities (5)

1

'unsafe-eval' found in script source

Security gaps expose your site and users to attacks, eroding trust.

Security › Content Security Policy
2

'unsafe-inline' found in script source

Security gaps expose your site and users to attacks, eroding trust.

Security › Content Security Policy
3

Cookie 'country' is missing the Secure flag

Security gaps expose your site and users to attacks, eroding trust.

Security › Cookie Security
4

Cookie 'locale' is missing the Secure flag

Security gaps expose your site and users to attacks, eroding trust.

Security › Cookie Security
5

Cookie 'oai-did' is missing the Secure flag

Security gaps expose your site and users to attacks, eroding trust.

Security › Cookie Security
View fix priority matrix

Fix Priority Matrix

5 findings

Quick Wins

5

High impact, low effort — start here.

Strategic

0

High impact, requires investment.

Nothing in this quadrant — good news.

Easy Improvements

0

Small gains, minimal effort.

Nothing in this quadrant — good news.

Deprioritize

0

Low impact, high effort — do last.

Nothing in this quadrant — good news.

← Low effort High effort →
BeaverCheck badge
Embed this badge
[![BeaverCheck](https://beavercheck.com/badge?url=https%3A%2F%2Fopenai.com)](https://beavercheck.com/results/660d8c5e-3b88-4564-930a-91985c5e0e51)
<a href="https://beavercheck.com/results/660d8c5e-3b88-4564-930a-91985c5e0e51"><img src="https://beavercheck.com/badge?url=https%3A%2F%2Fopenai.com" alt="BeaverCheck Score"></a>
https://beavercheck.com/badge?url=https%3A%2F%2Fopenai.com

This badge auto-updates with your latest scan result.

What fixing these means

Your site performs reasonably well, but a few targeted fixes could meaningfully improve results. Addressing the critical issues below would have the most immediate impact on your user trust.

5 security gaps detected — browsers may warn visitors about your site.

Conversion Barriers

3 critical 6 warning

9 barrier(s) likely increasing bounce by ~38%.

Speed (2)

Page takes 20.7s to load

+12% bounce

Users abandon at ~3s — you're 18.2s over the 2.5s threshold

Fix: Optimize render-blocking resources, preload the hero image, and compress images

Page feels frozen for 3.2s

+5% bounce

Clicks on the primary CTA are ignored while JavaScript runs

Fix: Break up long tasks; defer non-critical JavaScript to post-hydration

Trust (1)

No Content-Security-Policy header

+1% bounce

Higher XSS blast radius — one compromised script can exfiltrate the checkout form

Fix: Ship a reporting-only CSP first, then enforce once violations are clean

Usability (1)

No viewport meta tag

+15% bounce

Mobile browsers render at desktop width and shrink — text unreadable, tap targets miniature

Fix: Add <meta name="viewport" content="width=device-width, initial-scale=1">

Content (4)

No Open Graph tags

+2% bounce

Links shared on LinkedIn / Slack / Facebook show bare URLs — referral clicks drop

Fix: Add og:title, og:description, og:image, og:url to the page head

No structured data

+2% bounce

No rich-result eligibility in Google — lower SERP CTR vs competitors with stars and prices

Fix: Add JSON-LD for your page type (Product, Article, FAQPage, LocalBusiness, …)

Thin content

+3% bounce

Under 300 words — visitors bounce looking for substance, search engines rank competitors first

Fix: Add a substantive FAQ, product detail, or case-study section

Content is hard to read

+2% bounce

Graduate-level reading difficulty — limits the addressable audience for product or blog pages

Fix: Shorten sentences; replace jargon with plain language; target Flesch ease ≥60

Navigation (1)

No skip-to-content link

+1% bounce

Keyboard and screen-reader users must tab through the entire header on every page

Fix: Add a visible-on-focus <a href="#main">Skip to content</a> as the first focusable element

Preliminary CRO audit — each barrier links to the tab with detailed analysis.

Return on Investment

$250 investment → $0.89/month returns + USD 157,500 risk avoided

Payback period: > 2 years First-year ROI: -96%

Investment

$250

2h · 5 findings

Monthly returns

$0.89 /mo

~$11 / year

  • Bandwidth savings $0.89

Regulatory risk avoided

USD 157,500

if kept compliant

  • ADA Title III USD 150,000
  • CCPA/CPRA USD 7,500

Figures combine localized regulatory fine ceilings, search/conversion value priced against local CPC, and bandwidth waste estimates. Results depend on implementation quality and audience composition. Not legal or financial advice.

Full methodology & sources

Estimated Remediation Cost

$250

2.5 developer hours at $100/hr

Based on United States rates ($100/hr)

Quick wins
$250 5 fixes in ~150 minutes

Start here for the best return on investment

Cost by category

Cost by effort level

Adjust assumptions
$ /hr

Rates reflect fully-loaded developer cost including overhead

How developer rates are sourced

What Inaction Is Costing You

$13,126 / month at risk

~$157,511 / year if left unfixed

Compliance Risk

$157,500

ADA Title IIICCPA/CPRA
  • No privacy policy link detected
    CCPA/CPRA: USD 2,500 – USD 7,500
  • Trackers detected but no cookie policy found
    CCPA/CPRA: USD 2,500 – USD 7,500
  • Trackers detected but no privacy policy found
    CCPA/CPRA: USD 2,500 – USD 7,500

+6 more

Bandwidth Waste

$0.89 /mo

11088.7 MB/mo × 0.080 USD/GB

  • Optimize transfer: save ~1.1 MB per page load
    Saves $0.89/mo

Compliance figures represent the statutory maximum fine for the most severe triggered category, capped per regulation — not the sum of per-finding penalties. Based on published regulatory fine ranges. This is not legal advice.

Compliance methodology · SEO assumptions · Bandwidth model

Your performance is already good — improvements may show diminishing returns

Unique monthly visitors from your analytics

Purchases, signups, or key actions

Optional — for revenue estimation

additional conversions/month

more engaged visitors from reduced bounce

potential monthly revenue
Current bounce (est.)
After fixes (est.)
Estimated bounce reduction

Fix 5 critical issues to capture this value

How this is calculated

Based on Google/Deloitte research ("Milliseconds Make Millions") showing a ~7% bounce rate increase per additional second of LCP above the 2.5s "Good" threshold.

Your site's LCP: → estimated after fixes.

These are estimates based on industry research — actual results vary

Bounce-rate model & assumptions

Your data stays in your browser — nothing is sent to our servers

Was this report useful?

Thanks for your feedback!

We'll use a cached audit if available, or offer to scan.

Checking for existing audit...

Lighthouse Scores

Industry-standard audits powered by Google Lighthouse.

30
Performance Overall performance score (0–100) based on Core Web Vitals and other metrics. 90+ is good.
100
Accessibility Measures how accessible the page is for users with disabilities. Checks color contrast, ARIA labels, and semantic HTML.
81
Best Practices Checks for modern web development best practices including HTTPS, no console errors, and secure JavaScript.
100
SEO Measures basic SEO optimizations: meta tags, crawlability, link text, and mobile friendliness.

Core Web Vitals

Key metrics that affect user experience.

First Contentful Paint First Contentful Paint — how long until the browser renders the first piece of content. Under 1.8s is good.

4.37 s

Largest Contentful Paint Largest Contentful Paint — how long until the largest visible element loads. Under 2.5s is good.

20.67 s

Total Blocking Time Total Blocking Time — total time the main thread was blocked, preventing user input. Under 200ms is good.

3.20 s

Cumulative Layout Shift Cumulative Layout Shift — measures visual stability. How much the page layout shifts during loading. Under 0.1 is good.

0.000

Speed Index Speed Index — how quickly content is visually displayed during load. Under 3.4s is good.

8.01 s

Time to Interactive Time to Interactive — how long until the page is fully interactive and responds to user input. Under 3.8s is good.

21.77 s

Detailed Report

Audit breakdown by category with detailed findings.

30

Performance

Insights

Remove large, duplicate JavaScript modules from bundles to reduce unnecessary bytes consumed by network activity.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Polyfills and transforms enable older browsers to use new JavaScript features. However, many aren't necessary for modern browsers. Consider modifying your JavaScript build process to not transpile Baseline features, unless you know you must support older browsers. Learn why most sites can deploy ES6+ code without transpiling

Why this matters

Shipping ES5 transpiled code to modern browsers wastes bytes — every user with an evergreen browser pays for compatibility you don't need.

Learn more

Most users today run browsers that natively support ES6+, async/await, optional chaining, and the rest of modern JavaScript. Transpiling to ES5 'just in case' adds 20-40% to your bundle for no benefit. Configure your build to target a modern browserslist, or ship a differential bundle pair (modern + legacy) with the module/nomodule pattern.

Source: Google web.dev / Lighthouse

3rd party code can significantly impact load performance. Reduce and defer loading of 3rd party code to prioritize your page's content.

Why this matters

Performance issues directly impact user engagement and conversion rates.

A forced reflow occurs when JavaScript queries geometric properties (such as offsetWidth) after styles have been invalidated by a change to the DOM state. This can result in poor performance. Learn more about forced reflows and possible mitigations.

Why this matters

Performance issues directly impact user engagement and conversion rates.

headings: [map[key:source label:Source valueType:source-location] map[granularity:1 key:reflowTime label:Total reflow time valueType:ms]]
items: [map[reflowTime:49.208 source:map[type:text value:[unattributed]]]]

Each subpart has specific improvement strategies. Ideally, most of the LCP time should be spent on loading the resources, not within delays.

Why this matters

Performance issues directly impact user engagement and conversion rates.

headings: [map[key:label label:Subpart valueType:text] map[key:duration label:Duration valueType:ms]]
items: [map[duration:333.234 label:Time to first byte subpart:timeToFirstByte] map[duration:2318.537 label:Element render delay subpart:elementRenderDelay]]
lhId: page-0-P
nodeLabel: We use cookies to help this site function, understand service usage, and suppor…
path: 1,HTML,1,BODY,3,DIV,0,DIV,1,P
selector: body.text-primary-100 > div.bg-tertiary-80 > div.gap-4xs > p#cookieConsentDesc
snippet: <p id="cookieConsentDesc" class="text-p2 text-primary-60">

Avoid chaining critical requests by reducing the length of chains, reducing the download size of resources, or deferring the download of unnecessary resources to improve page load.

Why this matters

Performance issues directly impact user engagement and conversion rates.

description: [preconnect](https://developer.chrome.com/docs/lighthouse/performance/uses-rel-preconnect/) hints help the browser establish a connection earlier in the page load, saving time when the first request for that origin is made. The following are the origins that the page preconnected to.
title: Preconnected origins
description: Add [preconnect](https://developer.chrome.com/docs/lighthouse/performance/uses-rel-preconnect/) hints to your most important origins, but try to use no more than 4.
title: Preconnect candidates

A long cache lifetime can speed up repeat visits to your page. Learn more about caching.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Reducing the download time of images can improve the perceived load time of the page and LCP. Learn more about optimizing image size

Why this matters

Performance issues directly impact user engagement and conversion rates.

URLResource SizeEst Savings
1 1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/6eaq9qftNBJJjryJnZGVRj/991b778132f449626fb9...42.1 KiB38.8 KiB
oai Blog Codex Security Art Card 1x1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/5BrJdBcEkCtafmi26Ib87o/594a2f25a92633451971...20.9 KiB19.3 KiB
Learning Blocks ArtCard 1x1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/91H3WwMZWgZwhG890LVLR/b3da0d856865e08c37d78...15.0 KiB13.8 KiB
Amazon > Card div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/KnGPrCMI6RRSrpeLqMi5G/7cf83bc7e73df70949e3f...14.5 KiB13.4 KiB

These insights are also available in the Chrome DevTools Performance Panel - record a trace to view more detailed information.

Time to Interactive is the amount of time it takes for the page to become fully interactive. Learn more about the Time to Interactive metric.

Why this matters

Performance issues directly impact user engagement and conversion rates.

TTI

The maximum potential First Input Delay that your users could experience is the duration of the longest task. Learn more about the Maximum Potential First Input Delay metric.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Diagnostics

Reduce unused JavaScript and defer loading scripts until they are required to decrease bytes consumed by network activity. Learn how to reduce unused JavaScript.

Why this matters

Multi-megabyte JavaScript bundles delay every interactive feature on the page.

Learn more

This is the Lighthouse audit fired when too much JS is shipped relative to what executes. The fix isn't a config flag — it requires bundle analysis (webpack-bundle-analyzer, rollup-plugin-visualizer), splitting routes into chunks, lazy-loading off-screen components, and removing unused dependencies. Fundamentally different from minification: minifying reduces byte count, this reduces what's downloaded at all.

Source: Google web.dev / Lighthouse

Estimated savings: 9.2 s 1.8 MiB

Consider reducing the time spent parsing, compiling, and executing JS. You may find delivering smaller JS payloads helps with this. Learn how to reduce Javascript execution time.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Consider reducing the time spent parsing, compiling and executing JS. You may find delivering smaller JS payloads helps with this. Learn how to minimize main-thread work

Why this matters

Performance issues directly impact user engagement and conversion rates.

CategoryTime Spent
Script Evaluation4.8 s
Script Parsing & Compilation1.9 s
Other1.7 s
Style & Layout946 ms
Parse HTML & CSS203 ms
Garbage Collection166 ms
Rendering63 ms

More information about the performance of your application. These numbers don't directly affect the Performance score.

Layout shift culprits
Document request latency
Optimize DOM size
Font display
Modern HTTP
Optimize viewport for mobile
Minify CSS
Minify JavaScript
Reduce unused CSS
User Timing marks and measures 5 user timings
Avoid long main-thread tasks 20 long tasks found
Image elements have explicit `width` and `height`
Page didn't prevent back/forward cache restoration
Network Requests
Network Round Trip Times 10 ms
Server Backend Latencies 200 ms
Tasks
Diagnostics
Metrics
Screenshot Thumbnails
Final Screenshot
Script Treemap Data
Resources Summary
Avoid multiple page redirects
Initial server response time was short Root document took 320 ms
INP breakdown
LCP request discovery
Avoid non-composited animations
Avoid large layout shifts
100

Accessibility

These checks highlight opportunities to improve the accessibility of your web app. Automatic detection can only detect a subset of issues and does not guarantee the accessibility of your web app, so manual testing is also encouraged.

Visible text labels that do not match the accessible name can result in a confusing experience for screen reader users. Learn more about accessible names.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Failing Elements
Introducing GPT-5.4 Product 16 min read div.mb-sm > div.px-sm > div.group > a#4D3r5oL4eZTje3tZF6NhfE
Codex for (almost) everything Product 5 min read div.px-sm > div.relative > div.group > a#5bQ1R03qOM7xWLSWlXT9R6
The next evolution of the Agents SDK Product 5 min read div.px-sm > div.relative > div.group > a#3ffilEahAARsjx7Ez43BpR
Update on the OpenAI Foundation Company 3 min read div.px-sm > div.relative > div.group > a#2cp7NogxmNSfrgeeTZFDn8
Accelerating the next phase of AI Company Mar 31, 2026 div.max-w-container > div.group > div.group > a#6RvBi7O6laluCU3ymDSD5a
New ways to learn math and science in ChatGPT Product Mar 10, 2026 div.max-w-container > div.group > div.group > a#1oFzgkBotkTowN8dPwIIhl
Codex Security: now in research preview Product Mar 6, 2026 div.max-w-container > div.group > div.group > a#7gJt5XNwYeJTeQ5L6R9EjF
Our agreement with the Department of War Company Feb 28, 2026 div.max-w-container > div.group > div.group > a#5WZo3y5jpNICAbzXEylFEG
OpenAI and Amazon announce strategic partnership Company Feb 27, 2026 div.max-w-container > div.group > div.group > a#2og39hipbAUhN9b0uj3lqc
Improving instruction hierarchy in frontier LLMs Research Mar 10, 2026 div.@md:gap-sm > div.mb-md > div.group > a#7wu3OuFrpcwEV70Zt9duNx
Reasoning models struggle to control their chains of thought, and that’s good … div.@md:gap-sm > div.mb-md > div.group > a#43f38h660XvWgjSvmdiqRP
Extending single-minus amplitudes to gravitons Research Mar 4, 2026 div.@md:gap-sm > div.mb-md > div.group > a#1pFsNvcQVjGd11LEpMQ4T1
Gradient Labs gives every bank customer an AI account manager B2B Story Apr 1,… div.@md:gap-sm > div.mb-md > div.group > a#6nKM7pnc7VvuGFbnRZl1PB
STADLER reshapes knowledge work at a 230-year-old company B2B Story Mar 27, 20… div.@md:gap-sm > div.mb-md > div.group > a#7EBTmPbaeuvLCMNz0Zbaye
Rakuten fixes issues twice as fast with Codex B2B Story Mar 11, 2026 div.@md:gap-sm > div.mb-md > div.group > a#64hlvpwM49woI301TBA2XJ
English United States footer.mt-3xl > div.px-sm > div.relative > button#radix-_R_1tpdpsnpfivb_
Interactive controls are keyboard focusable
Interactive elements indicate their purpose and state
The page has a logical tab order
Visual order on the page follows DOM order
User focus is not accidentally trapped in a region
The user's focus is directed to new content added to the page
HTML5 landmark elements are used to improve navigation
Offscreen content is hidden from assistive technology
Custom controls have associated labels
Custom controls have ARIA roles
`[aria-*]` attributes match their roles
`[aria-hidden="true"]` is not present on the document `<body>`
`[role]`s have all required `[aria-*]` attributes
Elements with an ARIA `[role]` that require children to contain a specific `[role]` have all required children.
`[role]`s are contained by their required parent element
`[role]` values are valid
`[aria-*]` attributes have valid values
`[aria-*]` attributes are valid and not misspelled
Buttons have an accessible name
Image elements have `[alt]` attributes
Form elements have associated labels
`[user-scalable="no"]` is not used in the `<meta name="viewport">` element and the `[maximum-scale]` attribute is not less than 5.
ARIA attributes are used as specified for the element's role
Elements with `role="dialog"` or `role="alertdialog"` have accessible names.
`[aria-hidden="true"]` elements do not contain focusable descendents
Elements use only permitted ARIA attributes
Background and foreground colors have a sufficient contrast ratio
Document has a `<title>` element
`<html>` element has a `[lang]` attribute
`<html>` element has a valid value for its `[lang]` attribute
Links are distinguishable without relying on color.
Links have a discernible name
No element has a `[tabindex]` value greater than 0
Touch targets have sufficient size and spacing.
Heading elements appear in a sequentially-descending order
Document has a main landmark.
Deprecated ARIA roles were not used
Identical links have the same purpose.
`[accesskey]` values are unique
`button`, `link`, and `menuitem` elements have accessible names
ARIA input fields have accessible names
ARIA `meter` elements have accessible names
ARIA `progressbar` elements have accessible names
Elements with the `role=text` attribute do not have focusable descendents.
ARIA toggle fields have accessible names
ARIA `tooltip` elements have accessible names
ARIA `treeitem` elements have accessible names
The page contains a heading, skip link, or landmark region
`<dl>`'s contain only properly-ordered `<dt>` and `<dd>` groups, `<script>`, `<template>` or `<div>` elements.
Definition list items are wrapped in `<dl>` elements
ARIA IDs are unique
No form fields have multiple labels
`<frame>` or `<iframe>` elements have a title
`<html>` element has an `[xml:lang]` attribute with the same base language as the `[lang]` attribute.
Input buttons have discernible text.
`<input type="image">` elements have `[alt]` text
Lists contain only `<li>` elements and script supporting elements (`<script>` and `<template>`).
List items (`<li>`) are contained within `<ul>`, `<ol>` or `<menu>` parent elements
The document does not use `<meta http-equiv="refresh">`
`<object>` elements have alternate text
Select elements have associated label elements.
Skip links are focusable.
Cells in a `<table>` element that use the `[headers]` attribute refer to table cells within the same table.
`<th>` elements and elements with `[role="columnheader"/"rowheader"]` have data cells they describe.
`[lang]` attributes have a valid value
`<video>` elements contain a `<track>` element with `[kind="captions"]`
Tables have different content in the summary attribute and `<caption>`.
All heading elements contain content.
Uses ARIA roles only on compatible elements
Image elements do not have `[alt]` attributes that are redundant text.
Tables use `<caption>` instead of cells with the `[colspan]` attribute to indicate a caption.
`<td>` elements in a large `<table>` have one or more table headers.
81

Best Practices

General

Deprecated APIs will eventually be removed from the browser. Learn more about deprecated APIs.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Deprecation / WarningSource
SharedStorage
StorageType.persistent is deprecated. Please use standardized navigator.storage instead.
Fledge

Source maps translate minified code to the original source code. This helps developers debug in production. In addition, Lighthouse is able to provide further insights. Consider deploying source maps to take advantage of these benefits. Learn more about source maps.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Uses HTTPS
Avoids third-party cookies
Allows users to paste into input fields
Avoids requesting the geolocation permission on page load
Avoids requesting the notification permission on page load
Displays images with correct aspect ratio
Serves images with appropriate resolution
Page has the HTML doctype
Properly defines charset
No browser errors logged to the console
No issues in the `Issues` panel in Chrome Devtools
Ensure CSP is effective against XSS attacks
Ensure proper origin isolation with COOP
Mitigate DOM-based XSS with Trusted Types
Detected JavaScript libraries
Redirects HTTP traffic to HTTPS
Use a strong HSTS policy
Mitigate clickjacking with XFO or CSP
100

SEO

These checks ensure that your page is following basic search engine optimization advice. There are many additional factors Lighthouse does not score here that may affect your search ranking, including performance on Core Web Vitals. Learn more about Google Search Essentials.

Structured data is valid
Page isn’t blocked from indexing
Document has a `<title>` element
Document has a meta description
Page has successful HTTP status code
Links have descriptive text
Links are crawlable
robots.txt is valid
Image elements have `[alt]` attributes
Document has a valid `hreflang`
Document has a valid `rel=canonical`

Lighthouse Scores

Industry-standard audits powered by Google Lighthouse. — Desktop

56
Performance Overall performance score (0–100) based on Core Web Vitals and other metrics. 90+ is good.
97
Accessibility Measures how accessible the page is for users with disabilities. Checks color contrast, ARIA labels, and semantic HTML.
81
Best Practices Checks for modern web development best practices including HTTPS, no console errors, and secure JavaScript.
100
SEO Measures basic SEO optimizations: meta tags, crawlability, link text, and mobile friendliness.

Core Web Vitals

Key metrics that affect user experience. — Desktop

First Contentful Paint First Contentful Paint — how long until the browser renders the first piece of content. Under 1.8s is good.

527 ms

Largest Contentful Paint Largest Contentful Paint — how long until the largest visible element loads. Under 2.5s is good.

4.32 s

Total Blocking Time Total Blocking Time — total time the main thread was blocked, preventing user input. Under 200ms is good.

374 ms

Cumulative Layout Shift Cumulative Layout Shift — measures visual stability. How much the page layout shifts during loading. Under 0.1 is good.

0.012

Speed Index Speed Index — how quickly content is visually displayed during load. Under 3.4s is good.

2.54 s

Time to Interactive Time to Interactive — how long until the page is fully interactive and responds to user input. Under 3.8s is good.

4.55 s

Detailed Report

Audit breakdown by category with detailed findings.

56

Performance

Insights

Remove large, duplicate JavaScript modules from bundles to reduce unnecessary bytes consumed by network activity.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Polyfills and transforms enable older browsers to use new JavaScript features. However, many aren't necessary for modern browsers. Consider modifying your JavaScript build process to not transpile Baseline features, unless you know you must support older browsers. Learn why most sites can deploy ES6+ code without transpiling

Why this matters

Shipping ES5 transpiled code to modern browsers wastes bytes — every user with an evergreen browser pays for compatibility you don't need.

Learn more

Most users today run browsers that natively support ES6+, async/await, optional chaining, and the rest of modern JavaScript. Transpiling to ES5 'just in case' adds 20-40% to your bundle for no benefit. Configure your build to target a modern browserslist, or ship a differential bundle pair (modern + legacy) with the module/nomodule pattern.

Source: Google web.dev / Lighthouse

3rd party code can significantly impact load performance. Reduce and defer loading of 3rd party code to prioritize your page's content.

Why this matters

Performance issues directly impact user engagement and conversion rates.

A forced reflow occurs when JavaScript queries geometric properties (such as offsetWidth) after styles have been invalidated by a change to the DOM state. This can result in poor performance. Learn more about forced reflows and possible mitigations.

Why this matters

Performance issues directly impact user engagement and conversion rates.

headings: [map[key:source label:Top function call valueType:source-location] map[granularity:1 key:reflowTime label:Total reflow time valueType:ms]]
items: [map[reflowTime:51.007 source:map[column:84275 line:51 type:source-location url:https://openai.com/_next/static/chunks/07fsx~ixctd9k.js?dpl=dpl_E9iMCfvk7bFDDGxLZ88UH7DRNvX4 urlProvider:network]]]
headings: [map[key:source label:Source valueType:source-location] map[granularity:1 key:reflowTime label:Total reflow time valueType:ms]]
items: [map[reflowTime:197.616 source:map[type:text value:[unattributed]]] map[reflowTime:1.597 source:map[column:98170 line:50 type:source-location url:https://openai.com/_next/static/chunks/07fsx~ixctd9k.js?dpl=dpl_E9iMCfvk7bFDDGxLZ88UH7DRNvX4 urlProvider:network]] map[reflowTime:72.001 source:map[column:83712 line:50 type:source-location url:https://openai.com/_next/static/chunks/07fsx~ixctd9k.js?dpl=dpl_E9iMCfvk7bFDDGxLZ88UH7DRNvX4 urlProvider:network]]]

Reducing the download time of images can improve the perceived load time of the page and LCP. Learn more about optimizing image size

Why this matters

Performance issues directly impact user engagement and conversion rates.

URLResource SizeEst Savings
Amazon > Card div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/KnGPrCMI6RRSrpeLqMi5G/7cf83bc7e73df70949e3f...219.4 KiB217.4 KiB
Codex for (almost) everything div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/43j1ZX6CngGKjpcH3BRoaI/7228fda8953bf27c0c3c...98.2 KiB95.8 KiB
1 1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/6eaq9qftNBJJjryJnZGVRj/991b778132f449626fb9...98.1 KiB95.4 KiB
oai Blog Codex Security Art Card 1x1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/5BrJdBcEkCtafmi26Ib87o/594a2f25a92633451971...95.9 KiB95.0 KiB
A photograph of a woman using ChatGPT to help her work on a seed farm div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/4oc0vUwW6AaVePuwbMDODo/e8d1ed43aaaa8caa8769...121.0 KiB93.3 KiB
A photograph of a salvage yard that uses ChatGPT div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/7FEtIL0cIrDdIW5XGc2tCq/edf96f04ca7406977a96...115.9 KiB88.1 KiB
graviton paper release > card image div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/6GCHLf07ktUKcPeuulBOzk/0eba07515452c739fabb...89.8 KiB85.7 KiB
Frame div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/29yAmY0mMXwbU6gyRiHvGC/31703597eda0dc723d88...66.0 KiB65.4 KiB
Learning Blocks ArtCard 1x1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/91H3WwMZWgZwhG890LVLR/b3da0d856865e08c37d78...59.0 KiB58.4 KiB
OAIF Blogpost 1-1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/1369NNwJ3NhwM2y2U04sNA/9fcdfefd2e19135ac2e0...46.2 KiB45.0 KiB
Safety research the challenge > Cover Image div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/1dWPvCSG0V25TAlDws6dGU/214b389f0862c168692c...45.6 KiB43.5 KiB
A photograph of a family using ChatGPT to run their tamale shop div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/6JmoE3YM1XGL5Bozqcgs0p/f523f6726472d1361815...50.2 KiB36.1 KiB
our-agreement-with-the-department-of-war-1-1 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/1H3NKvKGj9aG4J3SFpBk3I/857ddd02725cc163f40d...28.3 KiB28.0 KiB
Agents SDK Art Card 1080x1080 div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/5TSaPxrSFnEbj3nQWfcSmY/661e54575d7318437e10...28.5 KiB27.8 KiB
"" div.size-full > div.group > picture.relative > img.object-cover
/kftzwdyauwt9/261kYlzPT8cBKoynLi35ct/14e3cae5d20ab9349af9...31.1 KiB23.4 KiB
CoT controllability > Cover Image div.relative > div.size-full > div.group > img.object-cover
/kftzwdyauwt9/7AkyH6nozJkoGaWK1VjBoj/adeeeeb93985405d4724...20.8 KiB19.9 KiB

Each subpart has specific improvement strategies. Ideally, most of the LCP time should be spent on loading the resources, not within delays.

Why this matters

Performance issues directly impact user engagement and conversion rates.

headings: [map[key:label label:Subpart valueType:text] map[key:duration label:Duration valueType:ms]]
items: [map[duration:341.769 label:Time to first byte subpart:timeToFirstByte] map[duration:2173.969 label:Element render delay subpart:elementRenderDelay]]
lhId: page-16-P
nodeLabel: We use cookies to help this site function, understand service usage, and suppor…
path: 1,HTML,1,BODY,3,DIV,0,DIV,1,P
selector: body.text-primary-100 > div.bg-tertiary-80 > div.gap-4xs > p#cookieConsentDesc
snippet: <p id="cookieConsentDesc" class="text-p2 text-primary-60">

Avoid chaining critical requests by reducing the length of chains, reducing the download size of resources, or deferring the download of unnecessary resources to improve page load.

Why this matters

Performance issues directly impact user engagement and conversion rates.

description: [preconnect](https://developer.chrome.com/docs/lighthouse/performance/uses-rel-preconnect/) hints help the browser establish a connection earlier in the page load, saving time when the first request for that origin is made. The following are the origins that the page preconnected to.
title: Preconnected origins
description: Add [preconnect](https://developer.chrome.com/docs/lighthouse/performance/uses-rel-preconnect/) hints to your most important origins, but try to use no more than 4.
title: Preconnect candidates

A long cache lifetime can speed up repeat visits to your page. Learn more about caching.

Why this matters

Performance issues directly impact user engagement and conversion rates.

These insights are also available in the Chrome DevTools Performance Panel - record a trace to view more detailed information.

Time to Interactive is the amount of time it takes for the page to become fully interactive. Learn more about the Time to Interactive metric.

Why this matters

Performance issues directly impact user engagement and conversion rates.

TTI

The maximum potential First Input Delay that your users could experience is the duration of the longest task. Learn more about the Maximum Potential First Input Delay metric.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Diagnostics

Reduce unused JavaScript and defer loading scripts until they are required to decrease bytes consumed by network activity. Learn how to reduce unused JavaScript.

Why this matters

Multi-megabyte JavaScript bundles delay every interactive feature on the page.

Learn more

This is the Lighthouse audit fired when too much JS is shipped relative to what executes. The fix isn't a config flag — it requires bundle analysis (webpack-bundle-analyzer, rollup-plugin-visualizer), splitting routes into chunks, lazy-loading off-screen components, and removing unused dependencies. Fundamentally different from minification: minifying reduces byte count, this reduces what's downloaded at all.

Source: Google web.dev / Lighthouse

Estimated savings: 1.7 s 1.8 MiB

Consider reducing the time spent parsing, compiling, and executing JS. You may find delivering smaller JS payloads helps with this. Learn how to reduce Javascript execution time.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Consider reducing the time spent parsing, compiling and executing JS. You may find delivering smaller JS payloads helps with this. Learn how to minimize main-thread work

Why this matters

Performance issues directly impact user engagement and conversion rates.

CategoryTime Spent
Script Evaluation1.3 s
Other493 ms
Script Parsing & Compilation487 ms
Style & Layout359 ms
Garbage Collection49 ms
Parse HTML & CSS40 ms
Rendering34 ms

More information about the performance of your application. These numbers don't directly affect the Performance score.

Layout shift culprits
Document request latency
Optimize DOM size
Font display
Modern HTTP
Optimize viewport for mobile
Minify CSS
Minify JavaScript
Reduce unused CSS
User Timing marks and measures 5 user timings
Avoid long main-thread tasks 13 long tasks found
Image elements have explicit `width` and `height`
Page didn't prevent back/forward cache restoration
Network Requests
Network Round Trip Times 10 ms
Server Backend Latencies 240 ms
Tasks
Diagnostics
Metrics
Screenshot Thumbnails
Final Screenshot
Script Treemap Data
Resources Summary
Avoid multiple page redirects
Initial server response time was short Root document took 330 ms
Avoid large layout shifts 1 layout shift found
INP breakdown
LCP request discovery
Avoid non-composited animations
97

Accessibility

These checks highlight opportunities to improve the accessibility of your web app. Automatic detection can only detect a subset of issues and does not guarantee the accessibility of your web app, so manual testing is also encouraged.

Best practices

Touch targets with sufficient size and spacing help users who may have difficulty targeting small controls to activate the targets. Learn more about touch targets.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Failing Elements
Research menu nav.nav-desktop:mask-[linear-gradient(to_right,black,black_calc(100%-1.5rem),transparent)] > ul.[&:has(>li:hover)>li:not(:hover)>*]:text-primary-60 > li.group/desktop-nav-item > button.group/desktop-nav-menu-button
Products menu nav.nav-desktop:mask-[linear-gradient(to_right,black,black_calc(100%-1.5rem),transparent)] > ul.[&:has(>li:hover)>li:not(:hover)>*]:text-primary-60 > li.group/desktop-nav-item > button.group/desktop-nav-menu-button
Business menu nav.nav-desktop:mask-[linear-gradient(to_right,black,black_calc(100%-1.5rem),transparent)] > ul.[&:has(>li:hover)>li:not(:hover)>*]:text-primary-60 > li.group/desktop-nav-item > button.group/desktop-nav-menu-button
Developers menu nav.nav-desktop:mask-[linear-gradient(to_right,black,black_calc(100%-1.5rem),transparent)] > ul.[&:has(>li:hover)>li:not(:hover)>*]:text-primary-60 > li.group/desktop-nav-item > button.group/desktop-nav-menu-button
Company menu nav.nav-desktop:mask-[linear-gradient(to_right,black,black_calc(100%-1.5rem),transparent)] > ul.[&:has(>li:hover)>li:not(:hover)>*]:text-primary-60 > li.group/desktop-nav-item > button.group/desktop-nav-menu-button

These items highlight common accessibility best practices.

Visible text labels that do not match the accessible name can result in a confusing experience for screen reader users. Learn more about accessible names.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Failing Elements
Introducing GPT-5.4 Product 16 min read div.mb-sm > div.px-sm > div.group > a#4D3r5oL4eZTje3tZF6NhfE
Codex for (almost) everything Product 5 min read div.px-sm > div.relative > div.group > a#5bQ1R03qOM7xWLSWlXT9R6
The next evolution of the Agents SDK Product 5 min read div.px-sm > div.relative > div.group > a#3ffilEahAARsjx7Ez43BpR
Update on the OpenAI Foundation Company 3 min read div.px-sm > div.relative > div.group > a#2cp7NogxmNSfrgeeTZFDn8
Accelerating the next phase of AI Company Mar 31, 2026 div.max-w-container > div.group > div.group > a#6RvBi7O6laluCU3ymDSD5a
New ways to learn math and science in ChatGPT Product Mar 10, 2026 div.max-w-container > div.group > div.group > a#1oFzgkBotkTowN8dPwIIhl
Codex Security: now in research preview Product Mar 6, 2026 div.max-w-container > div.group > div.group > a#7gJt5XNwYeJTeQ5L6R9EjF
Our agreement with the Department of War Company Feb 28, 2026 div.max-w-container > div.group > div.group > a#5WZo3y5jpNICAbzXEylFEG
OpenAI and Amazon announce strategic partnership Company Feb 27, 2026 div.max-w-container > div.group > div.group > a#2og39hipbAUhN9b0uj3lqc
Improving instruction hierarchy in frontier LLMs Research Mar 10, 2026 div.@md:gap-sm > div.mb-md > div.group > a#7wu3OuFrpcwEV70Zt9duNx
Reasoning models struggle to control their chains of thought, and that’s good … div.@md:gap-sm > div.mb-md > div.group > a#43f38h660XvWgjSvmdiqRP
Extending single-minus amplitudes to gravitons Research Mar 4, 2026 div.@md:gap-sm > div.mb-md > div.group > a#1pFsNvcQVjGd11LEpMQ4T1
Gradient Labs gives every bank customer an AI account manager B2B Story Apr 1,… div.@md:gap-sm > div.mb-md > div.group > a#6nKM7pnc7VvuGFbnRZl1PB
STADLER reshapes knowledge work at a 230-year-old company B2B Story Mar 27, 20… div.@md:gap-sm > div.mb-md > div.group > a#7EBTmPbaeuvLCMNz0Zbaye
Rakuten fixes issues twice as fast with Codex B2B Story Mar 11, 2026 div.@md:gap-sm > div.mb-md > div.group > a#64hlvpwM49woI301TBA2XJ
English United States footer.mt-3xl > div.px-sm > div.relative > button#radix-_R_1tpdpsnpfivb_
Interactive controls are keyboard focusable
Interactive elements indicate their purpose and state
The page has a logical tab order
Visual order on the page follows DOM order
User focus is not accidentally trapped in a region
The user's focus is directed to new content added to the page
HTML5 landmark elements are used to improve navigation
Offscreen content is hidden from assistive technology
Custom controls have associated labels
Custom controls have ARIA roles
`[aria-*]` attributes match their roles
`[aria-hidden="true"]` is not present on the document `<body>`
`[role]`s have all required `[aria-*]` attributes
Elements with an ARIA `[role]` that require children to contain a specific `[role]` have all required children.
`[role]`s are contained by their required parent element
`[role]` values are valid
`[aria-*]` attributes have valid values
`[aria-*]` attributes are valid and not misspelled
Buttons have an accessible name
Image elements have `[alt]` attributes
Form elements have associated labels
`[user-scalable="no"]` is not used in the `<meta name="viewport">` element and the `[maximum-scale]` attribute is not less than 5.
ARIA attributes are used as specified for the element's role
Elements with `role="dialog"` or `role="alertdialog"` have accessible names.
`[aria-hidden="true"]` elements do not contain focusable descendents
Elements use only permitted ARIA attributes
Background and foreground colors have a sufficient contrast ratio
Document has a `<title>` element
`<html>` element has a `[lang]` attribute
`<html>` element has a valid value for its `[lang]` attribute
Links are distinguishable without relying on color.
Links have a discernible name
Lists contain only `<li>` elements and script supporting elements (`<script>` and `<template>`).
List items (`<li>`) are contained within `<ul>`, `<ol>` or `<menu>` parent elements
No element has a `[tabindex]` value greater than 0
Heading elements appear in a sequentially-descending order
Document has a main landmark.
Deprecated ARIA roles were not used
Identical links have the same purpose.
`[accesskey]` values are unique
`button`, `link`, and `menuitem` elements have accessible names
ARIA input fields have accessible names
ARIA `meter` elements have accessible names
ARIA `progressbar` elements have accessible names
Elements with the `role=text` attribute do not have focusable descendents.
ARIA toggle fields have accessible names
ARIA `tooltip` elements have accessible names
ARIA `treeitem` elements have accessible names
The page contains a heading, skip link, or landmark region
`<dl>`'s contain only properly-ordered `<dt>` and `<dd>` groups, `<script>`, `<template>` or `<div>` elements.
Definition list items are wrapped in `<dl>` elements
ARIA IDs are unique
No form fields have multiple labels
`<frame>` or `<iframe>` elements have a title
`<html>` element has an `[xml:lang]` attribute with the same base language as the `[lang]` attribute.
Input buttons have discernible text.
`<input type="image">` elements have `[alt]` text
The document does not use `<meta http-equiv="refresh">`
`<object>` elements have alternate text
Select elements have associated label elements.
Skip links are focusable.
Cells in a `<table>` element that use the `[headers]` attribute refer to table cells within the same table.
`<th>` elements and elements with `[role="columnheader"/"rowheader"]` have data cells they describe.
`[lang]` attributes have a valid value
`<video>` elements contain a `<track>` element with `[kind="captions"]`
Tables have different content in the summary attribute and `<caption>`.
All heading elements contain content.
Uses ARIA roles only on compatible elements
Image elements do not have `[alt]` attributes that are redundant text.
Tables use `<caption>` instead of cells with the `[colspan]` attribute to indicate a caption.
`<td>` elements in a large `<table>` have one or more table headers.
81

Best Practices

General

Deprecated APIs will eventually be removed from the browser. Learn more about deprecated APIs.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Deprecation / WarningSource
SharedStorage
StorageType.persistent is deprecated. Please use standardized navigator.storage instead.
Fledge

Source maps translate minified code to the original source code. This helps developers debug in production. In addition, Lighthouse is able to provide further insights. Consider deploying source maps to take advantage of these benefits. Learn more about source maps.

Why this matters

Performance issues directly impact user engagement and conversion rates.

Uses HTTPS
Avoids third-party cookies
Allows users to paste into input fields
Avoids requesting the geolocation permission on page load
Avoids requesting the notification permission on page load
Displays images with correct aspect ratio
Serves images with appropriate resolution
Page has the HTML doctype
Properly defines charset
No browser errors logged to the console
No issues in the `Issues` panel in Chrome Devtools
Ensure CSP is effective against XSS attacks
Ensure proper origin isolation with COOP
Mitigate DOM-based XSS with Trusted Types
Detected JavaScript libraries
Redirects HTTP traffic to HTTPS
Use a strong HSTS policy
Mitigate clickjacking with XFO or CSP
100

SEO

These checks ensure that your page is following basic search engine optimization advice. There are many additional factors Lighthouse does not score here that may affect your search ranking, including performance on Core Web Vitals. Learn more about Google Search Essentials.

Structured data is valid
Page isn’t blocked from indexing
Document has a `<title>` element
Document has a meta description
Page has successful HTTP status code
Links have descriptive text
Links are crawlable
robots.txt is valid
Image elements have `[alt]` attributes
Document has a valid `hreflang`
Document has a valid `rel=canonical`

Send Feedback