01 — The client
Aquascapes Recreational Contractors has served Waterloo Region since 1994, building deep technical capability across vinyl-pool construction, seasonal service, maintenance, repairs and a retail store.
It is really several connected businesses — new builds, service, and retail — each with its own customers, decision stages and seasonal rhythm. None of that structure existed anywhere a machine could read it.
At a glance02 — The challenge
Pool construction, service & retail · Waterloo Region · seasonal demand
Traffic was declining and local competition was increasing. Three decades of technical capability and a strong local reputation were not visible to the systems people now use to choose a contractor.
The search environment had moved to Google Maps, search results, AI Overviews, Bing and AI assistants — surfaces that do not reward a reputation they cannot resolve. Nothing in the old presence told them what Aquascapes did, where, for whom, or on what evidence.
A machine cannot recommend what it cannot describe. Thirty years of proof that lives only in customers' heads is, to a search system, no proof at all.
The capability was real. The record of it was not.
03 — The strategy: an entity-first rebuild
Six blocks, in the order they had to happen. Each is a precondition for the next: you cannot rebuild a Business Profile against an entity model that does not exist yet, and you cannot expose reporting to an agent until there is reporting.
01
Modelled Aquascapes as it actually operates — legal identity, services, service areas, people, qualifications, certifications, technical knowledge, customer stories, project outcomes, store categories and seasonal workflows. That structure then shaped the content model, the information architecture, the booking flows and the structured data.
02
For a local business the profile is a major public connection point into Google's knowledge graph. Details, categories, services, attributes, service areas and content were rebuilt against the same entity model as the site, so the two agree instead of drifting.
03
A relational graph connecting the brand to its legal and operating identity — including the CRA business number — plus sameAs references, service categories, neighbourhoods, customer stories, qualifications and the evidence behind each claim. Shaun's expertise is modelled as knowsAbout relationships over vinyl pools, construction, maintenance, seasonal operation and troubleshooting.
04
A ground-up Astro build with React islands for the interactive journeys: a multi-step booking wizard for openings, closings and maintenance, seasonal and holiday-adjusted scheduling, Stripe payment collection, and neighbourhood pages tied to the customer stories that belong to them.
05
Google and Bing search setup, entity reconciliation, structured data and automated Business Profile posting. A Google Apps Script engine consolidates GA4, Search Console, Bing Webmaster Tools, Business Profile, Clarity and PageSpeed into one sheet.
06
A Cloudflare Worker exposes that reporting as an MCP server, so the numbers can be queried conversationally. The public site carries an API catalogue, an MCP server card, an A2A agent card and WebMCP tools — machine-readable ways for capable systems to discover and use what Aquascapes publishes.
04 — The site, at phone width
Astro with React islands for the interactive journeys. A multi-step booking wizard with seasonal and holiday-adjusted scheduling, Stripe payment, and neighbourhood pages tied to the customer stories that belong to them.
The live site, aquascapesrc.ca
The live site · aquascapesrc.ca
Lighthouse · 99 / 100 / 96 / 92
05 — What moved, and how we know
Every figure names the system it came from, so each one can be traced back to the place it was measured rather than taken on trust.
AI impressions
700%
The surfaces that answer rather than list — assistants, AI Overviews, agent queries — went from almost never naming Aquascapes to naming it routinely, on a record they could finally resolve.
Reporting engine, AI surfaces · baseline is the pre-rebuild presence
Figure
What moved
Window & source
450%
Website traffic, against the pre-rebuild baseline.
GA4 + Google Search Console
700%
AI impressions — the business being surfaced by assistants and AI answers rather than only by classic search.
Reporting engine, AI surfaces
4×
Retail store traffic and revenue, the part of the business the old site barely represented.
Client-reported retail figures
300%
New-build pipeline — enquiries for the highest-value service line.
Client-reported enquiries
99/100
Lighthouse Performance, with 100 Accessibility (AODA-grade), 96 Best Practices and 92 SEO. Independently re-runnable.
Lighthouse · anyone can re-run it
Not a new coat of paint. A business a machine can describe.
Get your AI Readiness Report ›More work
What happens next
Step 1 — See your score
Run a free check of how clearly AI can resolve, trust, and cite you, scored out of 100 across six signals. Your report lands in your inbox, usually within minutes.
Step 2 — Map the plan
We walk through your report, show you where AI hesitates, and map whether the Authority Foundation or the Authority Graph is the right system for your goal. Interpretation and planning, not a sales pitch.
