Core Web Vitals in 2026: What Actually Moves the Needle
INP replaced FID a while back and most sites still haven't adjusted their build habits for it. Here's what's actually worth fixing first.
Largest Contentful Paint still gets most of the attention because it's the easiest to explain, how long until the biggest thing on screen shows up. In practice, the bigger win for most sites in 2026 is Interaction to Next Paint. INP measures how responsive a page feels the moment someone actually tries to use it, tapping a menu, submitting a form, opening a panel, and a sluggish INP score quietly costs conversions in a way LCP alone won't catch.
The usual culprits are the same ones we find on almost every audit: unthrottled third-party scripts sitting in the critical path, JavaScript bundles that ship far more than the page needs, and render-blocking CSS or fonts delaying first paint for no real reason. We recently pulled a client's render-blocking icon font off the critical path entirely, deferring it to load after first paint instead, which alone shaved a meaningful chunk off their loading metrics with zero visual downside.
Image weight remains the single biggest lever on most marketing sites. Converting a page's images to WebP routinely cuts total payload by 70 to 90 percent with no visible quality loss, and it's often the fastest, lowest-risk fix available before touching a single line of application code.
None of this requires a rebuild. It requires actually measuring what's slow, fixing the two or three things responsible for most of the damage, and building future pages so those problems don't creep back in.
K4O5 is a digital studio in Upton, Northampton, building hand-coded websites, brand systems and cloud infrastructure for businesses across Northamptonshire, Milton Keynes and the Silverstone corridor.
Start a Project