Work we can prove.

The finding

Our 2026-07-09 audit hashed the raw HTML returned for every route on the site. All 19 URLs produced an identical MD5. The site was shipping a client-rendered shell with no server-side rendering and no prerendering, so the HTML a crawler received never varied by route — same title, same canonical, same body, on every page.

Every route initially returned the same app-shell HTML, forcing crawlers that rely on initial HTML — and Google before its rendering phase completes — to depend on JavaScript execution just to discover each route's unique content. Static prerendering moved the page-specific title, canonical, and body content into the initial response, removing that dependency entirely.

The fix

We moved the site onto Next.js prerendering so that each route emits its own fully-formed HTML document at build time. Every URL now ships a distinct <title>, a self-referencing <link rel="canonical">, and its own body content in the initial response — before any JavaScript runs.

The verification

We re-ran the identical audit on 2026-08-10 against the live production site. The previously reported technical SEO score was 35/100 on 2026-07-09 and 98/100 on 2026-08-10. The duplicate-HTML check that failed on the 19 audited URLs now passes on the same 19 URLs, confirmed by fetching each URL and diffing the raw response — not by trusting a dashboard.

Why this is the Gap Analysis pattern

The score measures the crawlability and indexability of the HTML a crawler receives. Between the two runs the site was also rebuilt on Next.js and its blog was reinstated, so we don't present this as a single-variable experiment: the check that failed was the duplicate-HTML check, and that is the check that now passes. That's the whole thesis: find what's structurally broken before you pay to drive traffic into it.

The finding

Production traffic was served as a static deployment — a file server reading pre-built HTML — while the site's own server-rendering code assumed a live server that never actually ran in production. Any logic that depended on that server — including server-side redirects — was silently doing nothing in production.

The fix

We rebuilt the site on a Vite + prerender pipeline so every route ships its own distinct <title>, canonical link, meta description, and structured data in the initial HTML response — fixing the identical-shell problem that made every URL indistinguishable to a crawler reading only the initial response. The page body still renders client-side after JavaScript runs; that part of the site was not part of this fix. Six legacy URLs from the old GoDaddy site were mapped to their new locations with static redirect stubs so old links and search-engine records land correctly instead of 404ing.

The verification

Per-route head metadata (title, canonical, description, structured data) confirmed live and correctly varying by URL, Sep 10 2026.

The finding

We installed GA4 on schkovl.com on 2026-08-26 behind a Consent Mode v2 banner: nothing loads until a visitor clicks Accept. For the next month the property showed "No data received". By 2026-09-10 we had ruled out a Content Security Policy block, then wrongly attributed the missing requests to the test browser's own extensions. A code audit on 2026-09-24 found the cause: the site's gtag stub pushed a plain array onto the dataLayer. gtag.js reads only Arguments objects from that queue, so every consent, config and page_view command was ignored — even after Accept.

The fix

The stub now pushes its Arguments object instead of a rest-parameter array: three lines added, two removed, in one file. A regression test asserts that each queued entry is an Arguments object, so the same mistake fails the test suite if it comes back.

The verification

Before the fix, a headless reproduction on 2026-09-24 sent zero /g/collect requests after accepting analytics. After the fix deployed on 2026-09-26, the queued dataLayer entries read [object Arguments], a page_view request went to /g/collect, and GA4 Realtime recorded first_visit, session_start and page_view for the test visit. On 2026-10-01 a fresh browser session sent no collect request before consent and two after clicking Accept.

Why we publish our own mistakes

A month with no data is exactly the kind of gap the Gap Analysis exists to find. The defect was invisible from the dashboard, survived a plausible wrong explanation, and fell to a code read that checked what the browser actually queued.