<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Yassir Msittef</title>
    <description>The latest articles on DEV Community by Yassir Msittef (@theghostdev90).</description>
    <link>https://dev.to/theghostdev90</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4256375%2Fba81caa8-379a-4c69-b6d1-4fe009aafd2b.JPG</url>
      <title>DEV Community: Yassir Msittef</title>
      <link>https://dev.to/theghostdev90</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/theghostdev90"/>
    <language>en</language>
    <item>
      <title>What 255 award-worthy websites taught me about Lighthouse</title>
      <dc:creator>Yassir Msittef</dc:creator>
      <pubDate>Fri, 20 Nov 2026 21:34:23 +0000</pubDate>
      <link>https://dev.to/theghostdev90/what-255-award-worthy-websites-taught-me-about-lighthouse-4eom</link>
      <guid>https://dev.to/theghostdev90/what-255-award-worthy-websites-taught-me-about-lighthouse-4eom</guid>
      <description>&lt;p&gt;I run &lt;a href="https://toprova.com/en" rel="noopener noreferrer"&gt;Toprova&lt;/a&gt;, a small web design award based in Geneva. Every site we review gets a human jury score (design, usability, creativity, content), and on top of that we run Lighthouse three times on the homepage and keep the median.&lt;/p&gt;

&lt;p&gt;After 255 measured sites, I had enough data to stop guessing. Here's what came out.&lt;/p&gt;

&lt;h2&gt;
  
  
  The headline numbers
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Median performance: 71/100.&lt;/strong&gt; Only 17% of sites reach 90 or more, and 5% are under 50.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Median accessibility: 91/100.&lt;/strong&gt; 58% reach 90+. Honestly better than I expected.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Median best practices: 96/100.&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Only 9% score 90+ on performance, accessibility and best practices at the same time.&lt;/strong&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;So the beautiful sites are mostly accessible and mostly well built. They're just not fast.&lt;/p&gt;

&lt;h2&gt;
  
  
  By category
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Category&lt;/th&gt;
&lt;th&gt;Median performance&lt;/th&gt;
&lt;th&gt;Median accessibility&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Portfolio&lt;/td&gt;
&lt;td&gt;83&lt;/td&gt;
&lt;td&gt;92&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;3D / WebGL&lt;/td&gt;
&lt;td&gt;79&lt;/td&gt;
&lt;td&gt;91&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Agency&lt;/td&gt;
&lt;td&gt;76&lt;/td&gt;
&lt;td&gt;92&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Brand&lt;/td&gt;
&lt;td&gt;73&lt;/td&gt;
&lt;td&gt;94&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Culture&lt;/td&gt;
&lt;td&gt;69&lt;/td&gt;
&lt;td&gt;90&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Event&lt;/td&gt;
&lt;td&gt;68&lt;/td&gt;
&lt;td&gt;90&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Editorial&lt;/td&gt;
&lt;td&gt;67&lt;/td&gt;
&lt;td&gt;91&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;E-commerce&lt;/td&gt;
&lt;td&gt;62&lt;/td&gt;
&lt;td&gt;93&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;The surprise for me: 3D / WebGL sites are &lt;em&gt;faster&lt;/em&gt; than e-commerce and editorial sites. My guess: heavy-looking sites are often built by people who obsess over every frame, while shops and magazines carry years of tags, widgets and third-party scripts. I haven't measured that part yet.&lt;/p&gt;

&lt;h2&gt;
  
  
  By stack
&lt;/h2&gt;

&lt;p&gt;Median performance by detected technology (at least 8 sites each):&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Technology&lt;/th&gt;
&lt;th&gt;Sites&lt;/th&gt;
&lt;th&gt;Median performance&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Astro&lt;/td&gt;
&lt;td&gt;12&lt;/td&gt;
&lt;td&gt;81&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;WebGL&lt;/td&gt;
&lt;td&gt;13&lt;/td&gt;
&lt;td&gt;78&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Lenis&lt;/td&gt;
&lt;td&gt;17&lt;/td&gt;
&lt;td&gt;76&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;GSAP&lt;/td&gt;
&lt;td&gt;20&lt;/td&gt;
&lt;td&gt;72&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;WordPress&lt;/td&gt;
&lt;td&gt;44&lt;/td&gt;
&lt;td&gt;71&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Nuxt&lt;/td&gt;
&lt;td&gt;16&lt;/td&gt;
&lt;td&gt;71&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Three.js&lt;/td&gt;
&lt;td&gt;20&lt;/td&gt;
&lt;td&gt;70&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Next.js&lt;/td&gt;
&lt;td&gt;55&lt;/td&gt;
&lt;td&gt;65&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Shopify&lt;/td&gt;
&lt;td&gt;13&lt;/td&gt;
&lt;td&gt;62&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Before anyone sharpens a pitchfork: Toprova itself is built with Next.js. I don't think this table says "framework X is slow". My guess is it says more about &lt;em&gt;what people build&lt;/em&gt; with each tool: big product and marketing sites with a lot of client-side JavaScript on one side, content-first sites on the other. Correlation, not a benchmark.&lt;/p&gt;

&lt;h2&gt;
  
  
  Caveats
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Desktop Lighthouse profile, homepage only.&lt;/li&gt;
&lt;li&gt;The sample is ambitious, design-heavy sites chosen by us or submitted by their authors. It's not the web at large.&lt;/li&gt;
&lt;li&gt;Lighthouse is a lab test, not real user data.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The full, live table is here: &lt;a href="https://toprova.com/en/barometre" rel="noopener noreferrer"&gt;Toprova web barometer&lt;/a&gt;. It updates every time the ranking does.&lt;/p&gt;

&lt;h2&gt;
  
  
  One more thing: can AI build a faster site than you?
&lt;/h2&gt;

&lt;p&gt;Everyone has an opinion on vibe coding. We'd rather see the builds. We just opened &lt;a href="https://toprova.com/en/build-fight" rel="noopener noreferrer"&gt;Build Fight&lt;/a&gt;: two creators get the same brief and a time limit (2 hours to 7 days), then the community votes blind. Use Cursor, Claude, v0, or no AI at all. Voters only see the result.&lt;/p&gt;

&lt;p&gt;If you want to be in the first vibe coder vs hand coder fight, join the queue.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: I built Toprova. I'd love to hear what you would measure instead of (or on top of) Lighthouse.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>webdev</category>
      <category>performance</category>
      <category>lighthouse</category>
      <category>design</category>
    </item>
  </channel>
</rss>
