Audit the actual website
Crawl public URLs and inspect status codes, redirects, canonical tags, sitemap entries and internal links. Compare source HTML with rendered pages and identify content that depends on fragile interactions. Review existing search data if the owner supplies access. A crawler report describes technical observations; it does not by itself prove ranking potential or a traffic forecast.
Map intent to useful pages
Group searches by the question or task they represent. Prefer one strong primary page for related commercial queries and use supporting guides for distinct informational questions. Geographic pages need meaningful local differences and truthful delivery information. Avoid creating interchangeable pages for every service, industry and city combination.
Entities, evidence and answers
Keep organization identity consistent and use structured data supported by visible content. Link service pages to relevant portfolio evidence and explain what that evidence actually demonstrates. Direct answers, clear headings and source references help readers and machine interpretation. No special file or schema guarantees inclusion in AI answers or a particular search position.
Validate and measure
Check changes against a saved baseline and verify the deployed site after release. Search Console can show query, page and country changes over time, but changes need context for seasonality and other releases. Test one hypothesis at a time when traffic supports it. Bring the domain, target customers and available search data to the audit discussion; rankings and backlinks are never fabricated.