UnitCycle — Product Context

Why This Exists

Property managers drown in data from Yardi but lack intelligence. UnitCycle transforms raw PM data into AI-driven insights, predictions, and automated workflows. The demo proves the concept with real Yardi-synced data (~15 properties, ~5,000 units, ~4,617 tenants, ~5,000 leases).

The Three Portals

PM Dashboard (Primary)

• Auth: admin / admin (mock credentials)
• Role: Alex Morgan, Finance Manager
• Experience: Dashboard-first with KPIs → drill into properties → manage units/tenants/leases
• Key pages: Dashboard, Properties, Tenants, Owners, Work Orders, Collections, Renewals, Invoices, Asset Health, Leasing Pipeline, Settings

Tenant Portal

• URL: /tenant-portal/login
• Auth: tenant / tenant (mock, maps to Brandon Cochran)
• Experience: Mobile-first with bottom nav, hero cards, 100dvh flexbox
• Key pages: Dashboard, Maintenance (submit/view), Payments (history/make), Lease, Chat, Profile

Maintenance Portal + Vendor Hub

• URL: /maintenance-portal/login
• Auth: TECH001 / 1234 (real API auth against maintenance_staff table)
• Techs: TECH001 = John Martinez, TECH002 = Maria Garcia
• Vendor Hub: /maintenance-portal/vendor-hub/jobs (job listing, bidding, my jobs)

UX Goals

• Dashboard-first design — every portal opens to KPIs and actionable items
• 5-star mobile app feel — bottom nav, no scrollbars, no dead-ends
• AI surfaces insights, not just data — predictions, recommendations, risk scores
• Navy + Gold branding — gold = ALL positive, red = danger only, NO semantic green
• Zero stock photos — real property images or none
• No mocked data — everything from real DB queries

Key User Journeys (with AI)

1. Lifecycle Tracking → auto-classify leases → state badges → action buttons per state
2. Renewals → churn scoring → rent strategy → AI-generated renewal letters
3. Collections → 5-step dunning → tone adaptation → payment plans → legal escalation
4. Work Orders → IoT alerts → auto-created WOs → vendor assignment → lifecycle tracking
5. Invoices → PDF upload → AI extraction → WO matching → price anomaly detection → approve/reject
6. Portfolio Chat → NL queries → streaming responses → inline tables → action buttons
7. Leasing Pipeline → screening → lease gen → move-in inspection → onboarding

Competitive Landscape

Benchmarked against: EliseAI, AppFolio, Buildium, Yardi, RealPage, BetterBot. Key differentiator: AI-first intelligence layer — proactive not reactive, revenue-first thinking, agents not dashboards, data flywheel.