{"id":18999665,"date":"2026-08-21T16:33:52","date_gmt":"2026-08-21T16:33:52","guid":{"rendered":"https:\/\/wordpress.org\/support\/topic\/powerful-plugin-but-the-ucss-feature-will-cost-you-time\/"},"modified":"2026-08-21T16:33:52","modified_gmt":"2026-08-21T16:33:52","slug":"powerful-plugin-but-the-ucss-feature-will-cost-you-time","status":"publish","type":"topic","link":"https:\/\/wordpress.org\/support\/topic\/powerful-plugin-but-the-ucss-feature-will-cost-you-time\/","title":{"rendered":"Powerful plugin, but the UCSS feature will cost you time"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">LiteSpeed Cache got our site to 95 mobile and 98 desktop on PageSpeed Insights. If you only looked at the numbers, you&#8217;d call it a five-star plugin. We&#8217;re giving it three, and the gap between those two facts is basically the whole review.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The core issue is Unique CSS. On paper it strips unused CSS per page and keeps things lean. In practice the Selector Allowlist isn&#8217;t fully trustworthy: a popup on our site quietly rendered broken for visitors for a stretch, and we didn&#8217;t catch it right away. Testing logged in as admin, everything looked fine. You have to test logged out, in a private session, every time, or you&#8217;re simply not seeing what your visitors see.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Even once you find the right fix, the caching behavior works against you. A correct CSS change can still look stale on the live page if you purge the page cache without also purging Unique CSS, or if the site isn&#8217;t freshly revisited before the next regeneration cycle runs. It&#8217;s easy to conclude a fix didn&#8217;t work when really it just hasn&#8217;t been given the chance to land yet.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">What makes this genuinely frustrating over time is that none of it stays fixed. A setup that&#8217;s been solid for weeks can quietly regress with no change on your end, often because another plugin (Elementor, in our case) updated its own markup or CSS selectors. LiteSpeed&#8217;s cache has no way of knowing that happened, so it just keeps serving a version that no longer matches. That&#8217;s not entirely LiteSpeed&#8217;s fault. It&#8217;s closer to a WordPress-wide problem: a site is a stack of independently updating plugins that have no awareness of each other, and LiteSpeed just happens to be the layer where that shows up first.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">So yes, the scores are real, and the plugin can genuinely deliver. But getting there took a lot of custom configuration and dev tools debugging, and staying there means you can&#8217;t treat it as solved and walk away. <br \/><br \/>We&#8217;re curious whether other people running Elementor on LiteSpeed have found a way to make these regressions less frequent, or whether periodic re-checking is simply the price of using it.<\/p>\n","protected":false},"template":"","class_list":["post-18999665","topic","type-topic","status-publish","hentry"],"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/wordpress.org\/support\/wp-json\/wp\/v2\/topic\/18999665","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wordpress.org\/support\/wp-json\/wp\/v2\/topic"}],"about":[{"href":"https:\/\/wordpress.org\/support\/wp-json\/wp\/v2\/types\/topic"}],"version-history":[{"count":0,"href":"https:\/\/wordpress.org\/support\/wp-json\/wp\/v2\/topic\/18999665\/revisions"}],"wp:attachment":[{"href":"https:\/\/wordpress.org\/support\/wp-json\/wp\/v2\/media?parent=18999665"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}