Internal build plan · private · Correlated Magnetics / PolyMagnet

Path B — a PolyMagnet site we host, off Shopify

Build a PolyMagnet site on our infrastructure so the automated marketing department can audit, draft and ship SEO, content and schema — measured against real leads and catalog orders. Keep the real store working the whole time.

Why do this

The engine can only prove itself on a real business with real leads and real commerce — dummy sites are "smoke and mirrors." PolyMagnet has years of lead and order history, so any lift is measurable and credible. Path B gives us a site we control end‑to‑end (so the department can inject changes on approval) without disturbing the live Shopify store until we've earned the switch.

Two ways to handle the catalog

Option 1

Hybrid — keep Shopify checkout

Rebuild the marketing surface (home, technology, applications, behaviors, company, guides) as a fast site we host, and keep the catalog + checkout on Shopify via its Buy Button / Storefront API, or as a linked /shop hook.

Pros: fastest to stand up; the proven commerce engine stays untouched; lowest risk.

Option 2

Full static rebuild

Export the Shopify catalog (products, variants, prices, images) to a static data file, rebuild the storefront and product pages, and wire checkout via Storefront API or Stripe.

Pros: maximum flexibility for the engine and speed. Cons: re‑implements tax/checkout/compliance — a bigger, later project.

Recommendation: start Option 1 (hybrid) — rebuild the marketing/content/SEO surface we control now, keep Shopify checkout. We get the case study fast and low‑risk; migrate fully to Option 2 later only if it clearly wins.

What we need from you

So we can build on a safe staging copy without touching the live store:

Shopify access — a staff invite (read is fine to start), or a product CSV export + image export.
Storefront API access token — only if we embed the live catalog/cart in the hybrid build.
A staging subdomain — e.g. new.polymagnet.com or try.polymagnet.com — so we build and test with zero risk to the live site. (You point one DNS record; we handle the rest.)
Brand assets — logo, product photography, any spec sheets or datasheets.
Analytics access (Google Analytics / Search Console) if available — so the "before" baseline is real.

How we build it

Stand up staging. A new.polymagnet.com on our droplet — zero risk to the live store.

Bring in the catalog. Import via CSV or Storefront API → product data + pages; wire checkout (hybrid: Shopify Buy Button / Storefront).

Rebuild the marketing surface. Home, technology, applications, behaviors, company, guides — with real SEO, schema and answer‑engine (AISO) structure. Agents‑First from day one.

Connect the department. The engine audits the staging site and drafts title/meta/schema/content/social — you approve each item, then it ships.

Build one‑click rollback. Every change is snapshotted; a single button instantly reverts to exactly how it was. Nothing publishes without a human approval.

Measure weekly. Traffic, inbound leads, catalog orders — tracked against the current baseline so the lift is undeniable.

Earn the switch. When staging out‑performs, cut polymagnet.com over by DNS — or keep the new site as the marketing front with Shopify running checkout.

Hosting, cost & safety

Hosting: our droplet + Backblaze backups — low, predictable cost.
Supervised model: the department drafts; a human approves; only then does it ship. Nothing auto‑posts to your accounts.
Rollback: snapshot‑before‑change + instant undo, so a bad change is never a real risk.
Tokens: AI usage tracked and invoiced against the $500 you authorized — kept lean.

Timeline

Staging + catalog import: a few days after we have access.
Marketing surface + department connected: ~1–2 weeks.
Target: something demoable by PCBC (end of July).

Path B · PolyMagnet build plan · the WholeTech team · private, not indexed