codelabs.ae — Rebuilding Our Own Search Foundation

The Real Problem Was Crawl Waste, Not Content

The instinct with 186 unindexed pages is to write more. That would have made it worse. The site was generating large numbers of thin taxonomy and archive pages — automatically created, near-identical, no reason for anyone to land on them — and Google was spending its budget discovering those instead of the pages that sell.

So the first move was subtractive: de-index the thin archives so crawl budget flows to real pages. The sitemap went from 17 sections to 6. Nothing of value was removed; a large amount of noise was.

Then The Unglamorous Part

Alt text on every image. Heading hierarchy corrected page by page. The eight broken internal links traced and repointed. Structured data added and validated. A machine-readable llms.txt published so AI answer engines have a clean description of what the company does, rather than inferring it.

Legacy URLs got individual attention rather than a blanket redirect. When the blog moved off the document root, old permalinks were left behind — and two of them turned out to be recoverable content rather than dead pages: one was a typo permalink missing a single letter, another was a post that had been re-slugged and never redirected. Both were pointed at the live articles. Pages with no equivalent went to the blog index, because sending a reader to an unrelated article is worse than telling them the truth.

What Changed

Measured in Google Search Console, comparing the three months of the work against the three months before it:

• Clicks: 390 → 882 — up 126%
• Impressions: 17.5K → 26.6K — up 52%
• Click-through rate: 2.2% → 3.3%
• Average position: 14.3 → 13.6

On the indexing side, which is what the crawl work was actually aimed at:

• Indexed pages: 92 → 112
• “Discovered – currently not indexed”: 102 → 22 — down 78%, and now passing validation
• Images missing alt text: 430 → 0 across the site’s primary pages
• Broken internal links: 8 → 0

What This Did Not Do

Worth being straight about, because it is the part most case studies leave out.

This was technical remediation. It made the site legible to search engines and got substantially more people to click through to it. It did not, on its own, turn that traffic into signed work — converting visibility into enquiries is a different piece of work, involving what the pages say and who they are aimed at, not whether Google can crawl them.

The honest summary is that we removed the reasons search engines had to ignore the site, and the traffic responded. Anyone promising you that a technical audit alone produces revenue is selling something else.

Why We Publish This One

Because it is the engagement where we had no ability to choose a flattering starting point. It was our own site, the baseline was genuinely poor, and the data is in a Search Console property we do not control the numbers of.

If you want the same audit pointed at your site, that is our SEO service.