The situation
Sparcford is an electrician based in Kingston upon Thames, serving South West London and Surrey. The site had been recently migrated from WordPress to Astro and was live with 12 pages: a homepage, three service pages, three blog posts, and a handful of utility pages.
Google Search Console showed the site was beginning to pick up impressions for relevant queries like "lighting installation kingston" (373 impressions) and "property rewiring in surrey" (251 impressions), but with an average position of 31.4 and only 3 clicks across 16 months, the site lacked the content depth to compete.
The goal: build a complete local SEO architecture that could rank for electrician keywords across Kingston, Richmond, Twickenham, Surbiton, and surrounding areas. Service pages, location pages, guide content, problem pages, and the technical SEO to tie it all together.
Our approach
01 Strategy and architecture design
We analysed the GSC data to understand what Google already associated the site with, reviewed the competitive landscape for local electrician searches, and designed a three-layer content architecture: service pages (money pages), location pages (geographic reach), and guide content (authority and long-tail traffic). Every page was planned with specific target keywords, internal linking relationships, and a clear role in the conversion funnel.
02 Service page buildout
We expanded from 3 service pages to 10, each targeting a specific service + location keyword pattern. Every page includes pricing guidance (Google and users both value this), an FAQ section targeting real search queries, and a sidebar with cross-links to related services. The existing rewiring and lighting pages were expanded with new content sections.
03 Location pages with genuine local content
11 location pages were created covering Kingston, Richmond, Twickenham, Surbiton, Teddington, Wimbledon, Hampton, Thames Ditton, Esher, Putney, and Wandsworth. Each page has genuinely unique content referencing local housing stock, property types, postcodes, and the specific electrical work common in that area. No templated doorway pages.
04 Guide and problem page content
12 blog posts were written: cost guides, safety guides, and "problem pages" targeting high-intent urgent searches like "fuse box keeps tripping" and "burning smell from socket". Each article links back to the relevant service page, passing authority to the money pages. Existing blog posts were rewritten and expanded.
05 Technical SEO and schema
FAQPage JSON-LD schema was added to all 9 service pages with FAQ sections (featured snippet eligibility). BreadcrumbList schema auto-generates on every page. The LocalBusiness schema was expanded with the full address, geo coordinates, and all 11 service areas. The footer was restructured with service links, area links, and company links for site-wide internal linking.
What was built
Service Pages
10
Consumer units, rewiring, fault finding, EICR, lighting, outdoor, sockets, kitchen, smart home, commercial
Location Pages
12
Hub + 11 individual area pages with unique local content
Blog Posts
12
Cost guides, safety guides, problem pages, and informational content
Schema Types
4
LocalBusiness, FAQPage, BreadcrumbList, WebPage
The result
Pages
Service Pages
Location Pages
Blog Posts
How AI made this possible
Building 41 pages of genuinely useful, unique content in a single session is not realistic with traditional methods. The AI workflow we used made it possible by handling the volume of content generation while we directed the strategy, reviewed the output, and managed the deployment pipeline.
Every page was reviewed on a Netlify staging branch before going live. The staging-first approach meant nothing reached production without being checked. The AI handled the content generation and code; we handled the architecture decisions, quality control, and deployment sequencing.
This is the kind of project that would typically take weeks. Not because the individual pages are complex, but because the sheer volume of content, internal linking, and technical SEO implementation is time-consuming when done manually. AI compressed the execution time without compromising the quality of the output.
Further reading
- What an AI system actually looks like — the pipeline architecture behind projects like this
- Start with one workflow, not a roadmap — how focused execution beats broad planning