🦡 BadgerFit

Find your perfect gym in Madison, WI — built for Badgers at CheesHacks 2025.

BadgerFit is a gym finder for the Madison, Wisconsin area. Users fill out a survey about their preferences (hours, budget, location, amenities, equipment, and fitness goals) and are matched with the top 3 gyms that best fit what they're looking for, along with a match percentage and the specific criteria that matched.

🚀 Getting Started Prerequisites

Node.js v18 or higher A Google Cloud account with the following APIs enabled:

Places API Geocoding API Maps JavaScript API

1. Clone the repo bashgit clone https://github.com/BenJPanackal/Badger-Fit.git cd Badger-Fit
2. Install dependencies bashnpm install
3. Set up your environment variables Create a .env file in the project root: GOOGLE_PLACES_KEY=your_google_api_key_here

⚠️ Never commit your .env file. It's already in .gitignore. Each person running the project needs to create their own .env with their own key, or use a shared key that the team agrees on.

4. Generate the gym data (run once) This script calls the Google Places API and builds gyms.json — the dataset that powers all matching. bashnode scripts/fetchGyms.js After it runs, open gyms.json and manually fill in the amenities, equipment, and monthlyPrice fields for each gym by checking their websites. This is what makes the matching meaningful.

✅ Once gyms.json is generated and filled in, commit it to the repo so teammates don't have to run this script themselves.

5. Start the server bashnode server.js
6. Open the app Go to http://localhost:3000 in your browser.

🗂️ Project Structure Badger-Fit/ server.js # Express server — entry point, run this to start the app gyms.json # Gym dataset generated by fetchGyms.js + manually enriched

routes/ match.js # POST /api/match — receives survey input, returns top 3 matches gyms.js # GET /api/gyms — returns full gym list

services/ matchService.js # Scoring algorithm — compares user preferences to gym data geocodeService.js # Converts a typed address to lat/lng via Google Geocoding API

scripts/ fetchGyms.js # One-time script to fetch all Madison gyms from Google Places

index.html # Home page survey.html # User preference survey output.html # Results page — displays top 3 matched gyms survey.js # Handles survey UI interactions survey.css # Styles for survey + output pages styles.css # Global styles for the rest of the site

🧠 How the Matching Works When a user submits the survey, their answers are sent to POST /api/match. The backend scores every gym in gyms.json across five weighted categories: CategoryWeightAmenities25%Equipment20%Hours20%Location20%Budget15% Each gym gets a score from 0–100 based on how many of the user's preferences it satisfies. The top 3 are returned with their match percentage and the specific criteria that matched (e.g. "Has pool", "Open during your preferred hours"). The user's fitness goal also influences equipment scoring — selecting "Build muscle" automatically factors in squat racks, free weights, and cable machines even if the user didn't check those boxes explicitly.

👥 Running on Multiple Machines Because gyms.json is committed to the repo, teammates only need to: bashgit clone https://github.com/BenJPanackal/Badger-Fit.git cd Badger-Fit npm install Then create their own .env file with a Google API key, and run: bashnode server.js Each person's server runs locally on their own machine on localhost:3000. The app does not require an internet-accessible server to run — it's fully local.

If you want the app accessible from other devices on the same WiFi network (e.g. for a demo), find your local IP address and share it: bash# Windows ipconfig

ifconfig Then visit http://YOUR_LOCAL_IP:3000 from another device on the same network.

🗺️ Roadmap ✅ Current Features

Survey-based gym matching with weighted scoring algorithm Match percentage + matched criteria displayed per gym Google Places integration for accurate gym data Distance scoring via Google Geocoding API

🔜 Planned: User Reviews We want to add a reviews feature where users can leave reviews on gyms that are visible to everyone using the app. This requires a shared backend database so reviews persist across sessions and machines. Recommended approach:

Add MongoDB Atlas (free tier) as the database — it's cloud-hosted so all users see the same data Install Mongoose: npm install mongoose Add a Review model with fields: gymPlaceId, author, rating, body, createdAt Add two new routes:

GET /api/reviews/:placeId — fetch all reviews for a gym POST /api/reviews — submit a new review

Render reviews on output.html below each matched gym card Add a MONGODB_URI variable to .env for the connection string

🛠️ Built With

Node.js + Express Google Places API Google Geocoding API Vanilla HTML, CSS, and JavaScript