# How to Vibe Code Your Own point.me (and Stop Paying for It)

> A better way to book with points

- Site: https://point.me
- Category: Travel / Rewards Metasearch
- Verdict: **Don't bother** (15/100 vibecodeable)
- Estimated effort: 6+ months of full-time work just to maintain basic scraper stability

## Verdict

Keep paying for point.me; building a personal clone is virtually impossible for a solo developer due to the insurmountable barrier of maintaining live, anti-bot-protected connections to 150+ airline reward systems.

Point.me relies on an extensive, brittle web of scrapers, direct partner integrations, and proprietary routing heuristics spanning over 150 airline systems that actively block automated traffic with sophisticated anti-bot walls (Cloudflare, Akamai, Imperva). While an AI coding agent can rapidly scaffold a clean Next.js frontend and a PostgreSQL database to mimic the UI and store simulated award data, it cannot magically bypass airline rate limits, emulate reverse-engineered loyalty API tokens, or keep up with continuous schema and award chart changes across 150 legacy carriers. You would spend every waking hour fixing broken scrapers rather than booking flights.

### What you can't replicate

- Live, unblocked query access to 150+ global airline and credit card reward program inventories
- Proprietary alliance partner ticketing rules and routing algorithms
- Established commercial affiliate agreements with card issuers

## What it does

Real-time metasearch and rewards optimization engine for finding and booking award flights using airline miles and credit card points across 150+ programs.

### Core features

- Real-time award flight search across 150+ loyalty and airline programs
- Side-by-side cash vs. credit card portal vs. transfer partner comparison
- Step-by-step points transfer instruction generation
- Portfolio tracker dashboard for loyalty balances ('My Points')
- Interactive deals map and price alert system ('Explore')
- Affiliate credit card integration and recommendation widgets

## The business

### Pricing

- Explore Free: Free — Limited destination map browsing and basic public deal feeds.
- Standard Membership: $12/mo or $129/yr — Full access to real-time award flight searches, point tracking, and transfer guides.
- Concierge Service: $200+ per passenger — White-glove human booking service for complex or high-value itineraries.

### Funding

$27M+ raised.
- Seed (July 2021): $2M
- Series A (September 2023): $10M
- Series B (September 2024): $15M
Investors: Citi Ventures, Nyca Partners, Thayer Ventures, Gaingels, Plug and Play Ventures, RiverPark Ventures, WndrCo, Brian Kelly (The Points Guy)

Founded 2019.
Team size: 10-50+.

## The hard parts

- Aggregating and scraping live, hidden award seat inventory across 150+ distinct, heavily protected airline backends
- Bypassing aggressive anti-bot protections (Cloudflare/Akamai) deployed by airline booking engines
- Maintaining low-latency search results under 20 seconds across dozens of concurrent upstream legacy systems
- Complex relational data mapping for dynamic award charts, alliance partner rules, and transfer partner constraints

## How to vibe code point.me

### Prerequisites

- Node.js (free): Required runtime environment for the Next.js frontend and scraper worker scripts.
- GitHub (free): Version control and CI/CD deployment pipeline integration.

### Recommended AI tools

- Claude Code: Best-in-class multi-file agentic coding tool for scaffolding the mock database schemas, UI dashboards, and scraper runner scripts.
- Cursor: Ideal for fine-tuning complex Tailwind CSS layouts, interactive deal maps, and side-by-side comparison tables.

### Stack

- Frontend: Next.js
- Backend: Next.js API Routes & TypeScript background workers
- Database: Neon
- Auth: better-auth
- Payments: Stripe
- Other: Tailwind CSS, Playwright, Vercel AI SDK

### Hosting

- Vercel (Hosting the Next.js frontend application and serverless API endpoints.): $0-20/mo
- Railway (Hosting persistent background worker containers required to execute simulated award search queues.): $5-10/mo

### Build guide

1. **Project Scaffolding & Database Schema** — Initialize the Next.js project with Tailwind CSS, configure TypeScript, and set up Neon Postgres tables for simulated award flights, user balances, and transfer partners.

```
Create a new Next.js project using Tailwind CSS and TypeScript. Configure a Neon Postgres database connection using Prisma or Drizzle ORM. Define the relational schema for users, loyalty_accounts (tracking point balances across Amex, Chase, airlines), award_flights (origin, destination, airline, cabin, points_required, cash_price, taxes, transfer_partner), and transfer_guides. Ensure all tables include proper foreign keys and indexes for fast searching by origin and destination airports. Implement better-auth for secure user email/password login and session handling.
```

2. **Portfolio Tracking Dashboard ('My Points')** — Build a user-facing dashboard that centralizes credit card rewards and airline miles balances in a single view.

```
Build a responsive user dashboard component in Next.js representing 'My Points'. Create UI cards for major reward currencies (Amex Membership Rewards, Chase Ultimate Rewards, Capital One Miles, Marriott Bonvoy, Delta SkyMiles, United MileagePlus). Allow users to manually input or update their balances, display total estimated cash value based on custom redemption cents-per-point multipliers, and provide a visual breakdown chart using Tailwind CSS and Lucide icons.
```

3. **Award Flight Search Engine & Comparison Matrix** — Develop the search input form and the side-by-side comparison matrix displaying cash, portal, and point transfer values.

```
Build an award flight search interface with inputs for origin airport, destination airport, departure date, passenger count, and cabin class (Economy, Business, First). Create a results page featuring a side-by-side comparison matrix. Each result card must display the airline, flight duration, nonstop vs. stops status, point cost with transfer breakdown (e.g., 57,500 points + $54 tax), equivalent cash price, credit card portal price, and a visual savings badge ('GOOD DEAL' or 'GREAT DEAL'). Populate the database initially with realistic mock search records to test rendering performance under load.
```

4. **Step-by-Step Transfer Guidance System** — Implement structured instructional flows showing how to transfer credit card rewards to specific airline partners.

```
Create a modal and dedicated view component for step-by-step points transfer instructions. When a user clicks 'Transfer to save X points' on any award flight search result, display a clear, multi-step wizard. Step 1: Log into your credit card rewards portal (e.g., Chase Ultimate Rewards). Step 2: Navigate to 'Transfer to Partners' and select the target airline (e.g., Air France Flying Blue) at a 1:1 transfer ratio. Step 3: Enter frequent flyer account details and confirm transfer. Step 4: Book the award seat directly on the airline website using transferred miles. Store transfer rules and partner ratios in the database for dynamic lookup.
```

5. **Interactive Deals Map & Explore View** — Build an interactive map interface to display top award sweet spots and flight deals worldwide.

```
Build an 'Explore' deals map page using Mapbox GL JS integrated into Next.js. Display map markers representing top global destinations (Paris, Tokyo, London, Sydney, etc.) with preview cards showing live promotional award redemption rates (e.g., Seattle to Tokyo for 25,750 points). Include filter drawers for point currency type, maximum point budget, and region to allow users to discover low-cost award sweet spots across flexible date ranges.
```

6. **Background Scraper Worker & Price Alerts** — Set up a background worker structure on Railway to periodically query or simulate airline inventory updates and trigger alerts.

```
Create a background worker service in TypeScript running on Railway that simulates polling or scraping airline award inventories on a cron schedule. Implement a price alert matching engine that compares newly indexed award rates against user-saved route watchlists. When a matching deal drops below a user's threshold, queue an outbound notification email using the Resend API with flight details and direct links back to the dashboard.
```

### Cost vs paying

**Starting costs (one-time):**

- Custom Domain: $12 one-time
- AI Coding Assistant (Claude Pro): $20 one-time
- Total: ~$32 one-time

**Ongoing costs (monthly):**

- Vercel Frontend Hosting: $20/mo
- Railway Background Worker & Database: $10/mo
- Resend Email Alerts: $0/mo (Free Tier)
- Total: ~$30/mo

- Paying for the SaaS instead: $12/mo (Standard Membership)
- Build time: 80-120 hours (Plus ongoing endless scraper maintenance)
- AI tool credits: $20 (Claude Pro)
- Break-even: Never (paying $12/mo is vastly cheaper and avoids the permanent maintenance trap of broken airline scrapers)

## Sources

- [Point.me Official Website](https://point.me)
- [PhocusWire: Point.me secures $15M Series B](https://www.phocuswire.com/point-me-secures-15m-series-b)
- [PitchBook: Point.me Company Profile](https://pitchbook.com/profiles/company/487313-27)