What a second actually costs
Speed is not a design preference, it is a revenue input. Google's research found that 53% of mobile visits are abandoned when a page takes longer than three seconds to load. Deloitte's Milliseconds Make Millions study measured an 8.4% lift in retail conversions from a 0.1-second improvement.
The uncomfortable part is that most sites do not know where they stand. They were fast on the developer's laptop and never measured again on a real phone, on real mobile data, with real third-party scripts loaded.
Where the time actually goes
Across the sites we audit, the same three culprits account for the overwhelming majority of wasted seconds.
Images nobody optimised
A single unoptimised hero image is routinely 2–4 MB. Served as AVIF with a WebP fallback, at the size it is actually displayed, that same image is often under 150 KB. Nothing else on this list gives you a bigger win for less effort.
Fonts blocking the first paint
Web fonts block text from rendering until the file downloads. Without font-display: swap, your visitor stares at a blank rectangle while a 90 KB font negotiates its way across a mobile connection.
Third-party scripts you forgot about
A chat widget, a heatmap tool, three analytics tags and an A/B testing script somebody installed two years ago and never removed. Each one is a render-blocking request that nobody owns and nobody reviews.
The three fixes that moved the needle
Serve modern image formats
Convert to AVIF with a WebP fallback. Set explicit width and height on every image so the browser reserves the space and the layout stops jumping. Lazy-load anything below the fold.
Preload the font, swap the fallback
Preload only the one weight used above the fold, and set font-display: swap. Text renders immediately in a system font, then swaps to your brand face once it arrives. Readers get content in milliseconds instead of staring at nothing.
Defer everything that isn't the page
Audit every third-party tag. Delete the ones nobody reads reports from. Defer the rest until after first paint. Your analytics does not need to load before your headline does.
How to measure it yourself
Run your busiest page through PageSpeed Insights and read the field data at the top, not the lab score underneath. Field data is what real visitors actually experienced over the last 28 days; the lab score is a simulation.
Largest Contentful Paint under 2.5 seconds is the threshold to clear. If you are above it, you are losing people who never told you why.
Where to start
Pick your highest-traffic page and fix the images first. It is almost always the single biggest improvement for the least work, and you can measure the difference the same afternoon.
If you would rather someone else did it, talk to us — performance work is most of what we do.
Written by · Founder & Software Developer
Shashi Bhushan Jha
Co-founder of FlikTech. Leads web engineering and delivery — performance, search visibility and the front-end craft that makes a site convert. Takes every kickoff call personally.
Want this applied to your business?
Reading about it is step one. Book a free consultation and we'll map it to your actual situation.
Start a conversation →