A 20-second load time is a revenue leak.
Performance audit and conceptual rebuild of a $2M+ revenue regional Vermont contractor's live website. All metrics independently measured with Lighthouse 12 and headless Chrome. Customer names anonymized.
Six measured failures.
These are not opinions. Each number below was captured against the live production URL on 2026-04-25 and is reproducible.
Largest Contentful Paint on a simulated 4G phone. Google's threshold for ‘Good’ is under 2.5s — the live site was over 8× the limit.
Desktop transfer size. Modern budget for a marketing site is ≈1.5 MB. Roughly 7.5× the healthy ceiling, mostly Wix editor runtime.
Wix bundles blocking the main thread. Lighthouse estimates 3.24 seconds of mobile load time spent parsing JS the page never executes.
Every service tile, hero CTA, and About-page label rendered as an <h1>. Destroys document outline for crawlers and screen readers.
Site declares <meta name="viewport" content="width=320"> — a legacy Wix hack that scales the desktop layout down instead of rendering a true mobile DOM.
No GeneralContractor / LocalBusiness JSON-LD anywhere in the site. The single largest leverage point for a regional contractor's local-pack ranking, missing entirely.
Same business, rewritten right.
A conceptual performance prototype built from the ground up on Next.js 16, statically rendered, modern image formats throughout.
Lighthouse 12 score on a throttled 4G phone — verified, not aspirational.
≈17× faster than the audited live site. Below Google's ‘Good’ threshold by more than half.
AVIF responsive source set, down from a 3.3 MB Wix-served JPG. Same visual fidelity, ≈99% lighter.
Name, email, phone, project type. The original asked for 14 fields including full street address before any conversation.
Before, after, delta.
Nine metrics, measured the same way on both sites.
Code is revenue infrastructure.
A 21-second mobile load time predicts roughly a 90% bounce rate for paid traffic. For a contractor running $2M+ in annual revenue, the gap between a 21-second and a 1-second site is not a UX preference — it is the difference between every cold lead seeing the brand and most of them never loading the page.
Re-architecting this regional leader's infrastructure on Next.js demonstrates that high-performance code is not a feature. It is a revenue protector. The same audit and the same rebuild process is what every Bravado HQ engagement runs.
Want this on your site?
Same audit. Same rebuild. Built around the way local customers actually find and trust your business.
This case study is an independent third-party performance audit and a conceptual rebuild prototype. It is not a representation of an active client engagement. All customer-identifying information has been anonymized. Live-site metrics were captured with publicly available tooling (Google Lighthouse 12, headless Chromium) on 2026-04-25.