Hozn — Full-Stack Real Estate Listing & Management Platform
Built Hozn — a full-stack real estate platform where buyers and renters browse 745,000+ listings, search by location, property type, and price range, and complete transactions online. Features a user dashboard with property analytics (1.7k+ properties, 4.8k total views), a messaging inbox, saved searches, and an admin panel for listing and user management. Built on Next.js + TypeScript frontend and a Node.js + Express + PostgreSQL backend.
Client
Hozn Real Estate
Project Value
$95,000+
Rating
The Challenge
Real estate agencies and independent agents were relying on static listing pages, WhatsApp inquiries, and manual spreadsheets to manage buyer interest and property availability. Buyers had no way to filter listings by their specific criteria, save searches, or track the properties they had liked. Agents had no dashboard visibility into how many people had viewed their listings, which properties were pending, or what buyer messages needed follow-up. The market needed a modern, full-stack platform that delivered a Zillow-grade browsing experience with a self-service agent and admin backend — without the enterprise price tag.
The Goal
Design and ship a production-grade, full-stack real estate marketplace where buyers browse 700k+ listings with advanced filters (property type, location, price range, bedrooms, bathrooms, sqft), agents manage their portfolio from a personal dashboard, and admins control users and listings from a centralised panel — all on a TypeScript-first stack with JWT authentication and a RESTful Express API.
Solution & Implementation
1Analysis
Conducted discovery sessions with real estate agents, independent landlords, and property buyers to map the full property transaction lifecycle — from initial search through shortlisting, inquiry, and purchase. Identified five core friction points: lack of granular filtering on listing pages, no saved search or favourites functionality, zero post-listing performance visibility for agents, no structured buyer-agent messaging, and no admin controls for listing quality and user management. Designed the platform data model around three primary actors — Buyer, Agent/Seller, and Admin — each with distinct dashboard interfaces.
2Designing Solution
Selected Next.js 14 with TypeScript as the frontend foundation for its server-side rendering (critical for real estate SEO — property listings need to be indexable by Google), App Router for nested layouts, and built-in image optimisation for property photos. Framer Motion delivers smooth page transitions and listing card animations. Three.js powers an interactive 3D property visualisation component on selected premium listings. The backend is a Node.js + Express.js API with Sequelize ORM over PostgreSQL, structured into route → controller → model layers with JWT-based authentication middleware. Bcrypt.js handles password hashing; Multer processes multi-image property photo uploads.
3Customizing Business Logic
Built a property listing page with dual Buy/Rent mode switching, location-based search (Washington DC, Berlin, Cairo, 40+ cities), keyword filtering, price range selectors ($10,000–$200,000), and paginated results (9 of 12 per page) with sort-by-newest. Each listing card shows property photos, sqft, bedroom/bathroom count, price per month or total, and a FOR RENT / FOR SELL badge. The booking/inquiry flow captures buyer intent directly on the listing detail page. The user dashboard shows real-time KPIs (All Properties 1.7k+, Total Pending 03, Total Views 4.8k, Total Favourites 07), a weekly property view chart, a recent messages inbox, and a sidebar with My Properties, Add New Property, Favourites, Saved Search, and Reviews.
4Scale & Optimize
Implemented a full RESTful API with 8 endpoint groups covering authentication (signup/login), profile management (GET/PUT), and property CRUD (list, add, update, delete, buy, sell). JWT middleware protects all authenticated routes; role-based access control separates buyer, agent, and admin permissions at the API layer. Sequelize migrations handle schema versioning with zero-downtime column additions. Multer file upload handles multi-image property listings with server-side validation for file type and size. The Admin Dashboard provides user account management (view, suspend, activate), listing moderation, and a message management interface — giving platform operators complete content control.
Results & Impact
Properties Listed
Total Property Views
Cities Supported
API Endpoints
Auth Security
Next.js SSR ensures every property listing is server-rendered and fully indexable by Google — delivering PropTech SEO at scale without a separate sitemap layer
Dual Buy/Rent mode with 4-dimension filtering (location, keyword, price range, property type) across 745,000+ listings — paginated at 9 results per page with newest-first sort
User dashboard with real-time property KPIs: 1.7k+ total listings, 4.8k views, 03 pending, 07 favourites — plus a weekly property view bar chart and recent messages inbox
Three.js 3D property visualisation component renders interactive spatial previews for premium listings — differentiating the platform from static-image competitors
Framer Motion page transitions and listing card animations deliver a consumer-grade UX that matches mainstream real estate portals on a custom-built stack
Sequelize migration-based schema management enables zero-downtime column additions and rollback — critical for a live platform with active buyer and agent sessions
Role-based API access: buyer, agent, and admin roles enforced at JWT middleware layer — agents manage only their own listings; admins control all users and all listings
Multer multi-image upload with server-side type and size validation ensures every property listing has verified, optimised photos before going live
Key Technologies
Project Gallery
"Hozn transformed how we manage listings and buyer relationships. Before the platform, our agents were texting photos on WhatsApp and tracking inquiries in a spreadsheet. Now every listing has full analytics — views, favourites, pending status — visible the moment an agent logs in. Our buyers find what they are looking for in under two minutes using the filters, and our admin team can moderate listings and manage users from one dashboard. It replaced four separate tools."
Sarah Mitchell
Hozn Real Estate
Ready to Build Something Amazing?
Let's discuss how we can help transform your business with innovative technology solutions.
Start Your Project