Lux Living Collective — Portal Tech Spec & Launch Plan (Revised)

Revision note: This replaces the Softr+Airtable approach. You're a software engineer with an existing Astro + Strapi + Cloudflare Workers stack. We build on what exists.

What You Already Have (and it's a lot)

Your existing site at ~/code/active/lux-property is a strong foundation:

Verdict: Don't start from scratch. Extend this.

Architecture: What to Add

EXISTING                           NEW (add these)
─────────────────────────────      ──────────────────────────────────
Astro SSR (Cloudflare Workers)     + Auth middleware (sessions via CF KV)
Strapi CMS                         + New content types (Clients, QuizResponses, Shortlists, Milestones)
React components                   + Portal dashboard components
Landing pages (30 variants)        + 3 segment pages (investors, downsizers, FHB)
Lead capture forms                 + Apartment matching quiz
n8n webhooks                       + Email nurture triggers
PostHog                            + Portal event tracking (funnel analytics)
Meta CAPI                          + Quiz completion events
Phone verification                 + (reuse for portal signup)

What You DON'T Need

• No separate portal app (build it into the existing Astro site)
• No Softr/Airtable ($0/month saved)
• No separate auth provider (Cloudflare Workers + KV handles sessions)
• No separate community platform yet (start with WhatsApp or a simple /community page)
• No separate email platform initially (n8n can trigger emails via Resend/Postmark API)

Phase 1: Seal the Funnel (Week 1-2)

This is the highest-impact work. Everything in the repositioning strategy depends on this.

1A. New Pages (Astro)

Create these pages in the existing site:

src/pages/
├── quiz/
│   └── index.astro              ← Apartment matching quiz (React component)
├── book/
│   └── index.astro              ← Consultation booking (Calendly embed or custom)
├── how-it-works/
│   └── index.astro              ← Advisory process explainer
├── for-investors/
│   └── index.astro              ← Investor-focused landing page
├── for-downsizers/
│   └── index.astro              ← Downsizer-focused landing page
├── first-home-buyers/
│   └── index.astro              ← FHB-focused landing page
├── guide/
│   └── index.astro              ← Buyer's guide download (email gate)
└── api/
    ├── quiz-submit.ts           ← Quiz submission → Strapi + n8n
    └── guide-download.ts        ← Guide request → email capture + n8n

Copy: Use the landing page copy from mvp/LANDING-PAGES.md — it's already written.

1B. Apartment Matching Quiz (React Component)

A multi-step form component that lives at /quiz/. This is your primary lead magnet.

src/components/react/Quiz/
├── Quiz.tsx                     ← Main quiz container (step management)
├── QuizStep.tsx                 ← Individual question step
├── QuizResults.tsx              ← "We found 3 matches" teaser + booking CTA
└── quiz-config.ts               ← Questions, options, scoring logic

Quiz flow:

1. 5 questions (reason for buying, area preference, budget, priorities, timeline)
2. Animated step transitions
3. Results page: "Based on your answers, we've matched you with 3 developments"
4. NO project names shown — just attributes ("a boutique development in the Inner West with rooftop amenities")
5. CTA: "Book a free 15-min call to get your personalised shortlist"
6. Capture: name, email, phone
7. Submit → Strapi (new QuizResponse content type) + n8n webhook (triggers email nurture)

Quiz data goes to Strapi as a new content type:

QuizResponse:
  - name: string
  - email: string (unique)
  - phone: string
  - reason: enum (first-home, upgrade, downsize, investment)
  - areas: json (array of selected areas)
  - budget: enum (under-800k, 800k-1.2m, 1.2m-2m, 2m-plus)
  - priorities: enum (location, design, investment, lifestyle)
  - timeline: enum (now, 3-6-months, 6-12-months, researching)
  - source: string (utm_source, utm_campaign tracking)
  - status: enum (new, contacted, consultation-booked, active-client, purchased)
  - createdAt: datetime

1C. Homepage Rework

Update the existing homepage (src/pages/index.astro) to match the repositioning:

Current: "Elevate Your Living" + 6 named project cards
New: "The Smart Way to Buy a New Apartment in Sydney" + How It Works + quiz CTA

You don't need to delete the project pages yet — just change what the homepage leads with. The project pages can stay but be de-emphasised (remove from nav, keep for SEO).

1D. API Routes

Add to existing src/pages/api/:

// src/pages/api/quiz-submit.ts
// POST: receives quiz answers + contact info
// → Creates QuizResponse in Strapi
// → Sends to n8n webhook (triggers email nurture sequence)
// → Fires Meta CAPI event (Lead / CompleteRegistration)
// → Returns success + redirect to /book/

// src/pages/api/guide-download.ts
// POST: receives name + email
// → Creates or updates contact in Strapi
// → Sends to n8n webhook (triggers guide email + nurture)
// → Returns PDF URL or sends via email

Phase 1 Effort Estimate

Monthly cost: $0 additional (runs on existing Cloudflare Workers plan)

Phase 2: Client Portal (Week 3-5)

Once you're capturing leads through the quiz, build the portal that keeps them in your ecosystem.

2A. Authentication

Use Cloudflare Workers KV for session management. No need for Auth0/Clerk/Supabase.

src/
├── middleware.ts                 ← Astro middleware: check session cookie, protect /portal/* routes
├── pages/
│   ├── login/
│   │   └── index.astro          ← Magic link login (email-based, no password)
│   ├── portal/
│   │   ├── index.astro          ← Dashboard home
│   │   ├── shortlist.astro      ← Their matched developments
│   │   ├── checklist.astro      ← Buying journey checklist
│   │   ├── resources.astro      ← Guides, documents, links
│   │   └── profile.astro        ← Their preferences and contact info
│   └── api/
│       ├── auth/
│       │   ├── login.ts         ← Send magic link email
│       │   ├── verify.ts        ← Verify magic link token, create session
│       │   └── logout.ts        ← Destroy session
│       └── portal/
│           ├── shortlist.ts     ← CRUD shortlist items
│           └── checklist.ts     ← Update checklist progress

Why magic links:

• Zero friction (no password to remember)
• Reuses the phone verification pattern you already have (but via email)
• Feels premium (like high-end services)
• You already have their email from quiz/consultation

Session flow:

1. Client enters email on login page
2. API generates a token, stores in KV with 15-min TTL, sends email via Resend/Postmark
3. Client clicks link → API verifies token → creates session cookie → redirects to /portal/
4. Session stored in Cloudflare KV (key: session token, value: client ID, TTL: 30 days)
5. Astro middleware checks session cookie on every /portal/* request

2B. Strapi Content Types (Portal Data)

Extend your existing Strapi with:

Client (new content type):
  - name: string
  - email: string (unique)
  - phone: string
  - segment: enum (investor, downsizer, fhb, upgrader)
  - status: enum (lead, quiz-completed, consultation-booked, active, purchased, post-purchase)
  - quizResponse: relation (→ QuizResponse)
  - shortlist: relation (→ Project[], many-to-many)
  - notes: richtext (your internal notes about this client)
  - consultationDate: datetime
  - assignedTo: string (which team member is managing them)
  - source: string (utm tracking)
  - createdAt: datetime

ClientMilestone (new content type):
  - client: relation (→ Client)
  - milestone: enum (joined, quiz-completed, consultation-booked, pre-approved,
                      shortlist-received, display-suite-visited, offer-made,
                      contract-signed, settlement-complete)
  - completedAt: datetime
  - notes: text

Shortlist (new content type):
  - client: relation (→ Client)
  - project: relation (→ Project)
  - yourNotes: richtext (why you recommended this one)
  - status: enum (recommended, interested, visited, rejected, chosen)
  - addedAt: datetime

Key insight: Projects already exist in Strapi with full details (developer, architect, builder, amenities, images, pricing). The Shortlist content type just creates a client-specific view of those projects with your advisory notes attached. The client sees the projects you've curated for them — not the full database.

2C. Portal Dashboard (React Components)

src/components/react/Portal/
├── Dashboard.tsx                ← Overview: stage indicator, next steps, recent updates
├── ShortlistView.tsx            ← Cards showing their matched developments
├── ProjectCard.tsx              ← Individual project in shortlist (uses existing StrapiProject type)
├── JourneyChecklist.tsx         ← Buying journey stages with checkboxes
├── ResourceLibrary.tsx          ← Guides, documents, useful links
└── ProfileSettings.tsx          ← Edit preferences, contact info

Dashboard shows:

• Where they are in the journey (visual progress bar)
• Their shortlisted developments (revealed after consultation — this is the funnel seal)
• Next recommended action ("Book your first display suite visit" / "Review the contract checklist")
• Upcoming Q&A sessions
• Direct message link to their advisor (you)

2D. The Funnel Seal in Action

Here's how the portal completes the repositioning strategy:

1. Lead takes quiz → captured in Strapi, no project names revealed
2. Lead books consultation → you discuss needs, recommend 2-3 projects
3. You create Shortlist entries in Strapi → link Client to Projects with your notes
4. Client gets portal access (magic link email)
5. Client logs in → sees their personalised shortlist with YOUR context
   - "We recommend this because..."
   - "The developer has a strong track record..."
   - "This is undervalued compared to..."
6. Client can mark projects as interested/visited/rejected
7. You see their engagement in Strapi admin → know when to follow up

They CAN still Google the project name at this point, but:
- You've already built trust through the consultation
- Your advisory notes add value they can't get from the developer
- The portal gives them a reason to stay in your ecosystem
- They see you as THEIR advisor, not the project's salesperson

Phase 2 Effort Estimate

Monthly cost: ~$5-10/month (Resend for transactional emails, everything else on existing CF plan)

Phase 3: Email Nurture & Automation (Week 4-6)

3A. Email Provider

Existing: Brevo (already in place) — use for now or migrate to Plunk (open-source, self-hostable, clean API for transactional + automated sequences).

Plunk handles nurture sequences natively — no need to route email through n8n. Keep n8n for what it's already doing (form webhooks, other automations). Email delivery and sequencing live directly in your email platform.

Quiz submit → API route:
  1. Save QuizResponse to Strapi
  2. Call Plunk/Brevo API: create contact, tag by segment, trigger sequence
  3. Fire Meta CAPI event
  4. Return success

3B. Email Sequences

Use the 19 emails from mvp/QUIZ-AND-EMAIL.md. They're already written. Set up these sequences in Plunk/Brevo directly:

Sequences (configured in Plunk/Brevo):
├── quiz-nurture-investor (5 emails over 14 days)
├── quiz-nurture-downsizer (6 emails over 21 days)
├── quiz-nurture-fhb (6 emails over 21 days)
├── consultation-booked (confirmation + prep email)
├── post-consultation (shortlist ready + portal invite)
└── no-show-reengagement (2 emails)

Trigger: API call from quiz-submit or consultation-booked API route → Plunk/Brevo adds contact to the right sequence based on segment tag.

3C. Portal Notifications

Once the portal exists, add email triggers for portal events:

• "Your shortlist is ready — log in to view" (after you create their shortlist)
• "New development added to your shortlist" (when you add a match)
• "Monthly Q&A this Thursday — submit your questions" (community)
• "Market update for [their preferred area]" (content)

These are transactional emails sent via Plunk/Brevo API — not sequences.

Phase 3 Effort Estimate

**Monthly cost: $0-25/month** (Brevo free tier or Plunk self-hosted, or Plunk Cloud ~$25/mo)

Phase 4: Community (Month 2-3)

Start with the simplest thing that works. Graduate complexity as membership grows.

Option A: WhatsApp Community (Recommended for launch)

Zero code. See COMMUNITY-LAUNCH-PLAN.md for the full playbook. Add a /community page to the site with:

• What the community is
• What happens there (Q&A sessions, market updates, peer support)
• Join link (WhatsApp invite)
• Screenshots/social proof

Option B: Simple Forum Page (When you outgrow WhatsApp)

Build a basic discussion space into the Astro site. Use Cloudflare D1 (SQLite at the edge) as the database.

src/pages/community/
├── index.astro                  ← Topic list
├── [topic].astro                ← Thread view
├── new.astro                    ← New topic form (auth required)
└── api/
    ├── topics.ts                ← CRUD topics
    └── replies.ts               ← CRUD replies

D1 Schema:
  topics: id, title, author_id, category, body, created_at, last_reply_at, reply_count
  replies: id, topic_id, author_id, body, created_at
  categories: id, name, slug, description

This is intentionally simple. No notifications, no reactions, no threading. Just topics and replies. You can add features as the community grows, or migrate to Flarum/Discourse if it takes off.

Option C: Flarum (If community becomes core to the business)

Self-host on a $18/month DigitalOcean droplet at community.luxlivingcollective.au. Only worth it if you hit 100+ active members and need moderation tools, search, notifications, mobile PWA.

Recommended path: A → B → C as the community grows.

Full Cost Summary

Tech Stack Summary

FRONTEND        Astro 5.14 + React (existing) + Tailwind 3.4 (existing)
RUNTIME         Cloudflare Workers (existing) + Bun (existing, local dev)
CMS/DATABASE    Strapi (existing) — extended with Client, QuizResponse, Shortlist, Milestone types
AUTH            Magic links + Cloudflare KV sessions (new, ~6hrs to build)
EMAIL           Plunk or Brevo (existing) — sequences + transactional, no n8n needed for email
ANALYTICS       PostHog (existing) + Meta CAPI (existing)
COMMUNITY       WhatsApp initially → built-in forum later → Flarum if needed
FILE STORAGE    Cloudflare R2 (if needed for guide PDFs, documents)

Build Sequence

Sprint 1 (Week 1-2): Seal the Funnel

Priority: Stop the leak. Get quiz live, get segment pages live, update homepage.

1. Create Strapi QuizResponse content type
2. Build quiz React component
3. Build /quiz/ page + /api/quiz-submit route
4. Create 3 segment landing pages (/for-investors/, /for-downsizers/, /first-home-buyers/)
5. Create /how-it-works/ and /book/ pages
6. Update homepage hero + navigation
7. Set up n8n workflow: quiz submit → email notification to you
8. Deploy

Sprint 2 (Week 3-4): Email Nurture

Priority: Don't let quiz leads go cold. Automated follow-up.

1. Set up Plunk (or configure Brevo) — API integration from API routes
2. Build HTML email templates (base layout + 3-4 variants)
3. Create email sequences in Plunk/Brevo (investor, downsizer, FHB)
4. Wire quiz submission → segment detection → correct sequence
5. Add Meta CAPI events for quiz completion
6. Create /guide/ page + buyer's guide PDF

Sprint 3 (Week 5-7): Client Portal

Priority: Give consultation clients a reason to stay.

1. Create Strapi Client, Shortlist, ClientMilestone content types
2. Build magic link auth (login page, API routes, KV sessions, middleware)
3. Build portal dashboard page + React components
4. Build shortlist view (shows projects you've recommended)
5. Build journey checklist
6. Wire portal invite email (sent after consultation)
7. Deploy

Sprint 4 (Week 8-10): Community + Polish

Priority: Start building the peer network.

1. Launch WhatsApp community (zero code, see community launch plan)
2. Add /community/ page to site (info + join link)
3. Run first Q&A session
4. Add Google Ads campaigns (search intent keywords from repositioning strategy)
5. A/B test ad creatives (use the 22 ad units from mvp/AD-CREATIVES.md)
6. Track full funnel: ad → quiz → consultation → portal → purchase

What We Build Together

Given you're a software engineer and I'm here to help, here's how we'd split the work:

When you're ready to start building, we can work through each sprint together. I can write the code, you review and deploy. The quiz component alone would seal most of the funnel leak and could be live in a day or two.

Success Metrics (Same as Before)

This approach costs ~$20/month instead of $176/month, gives you full control over the codebase, builds on infrastructure you already maintain, and can be built incrementally starting with the highest-impact piece (the quiz that seals the funnel).