Goodsline

Inspiration

We wanted to create a tool that makes sense of the overwhelming amount of digital chatter around products. Inspired by the noise-to-signal gap in social media, Goodsline emerged from the Perplexixity Hackathon as a solution to decode public sentiment and spotlight what truly matters—how people feel about products.

What it does

Goodsline analyzes product-related conversations on YouTube and Reddit using Perplexity Sonar. It extracts comments, applies custom sentiment scoring over five years of data, and calculates a dynamic relevancy score based on recent user engagement. The results are visualized in a sleek, intuitive dashboard for real-time insights.

How we built it

We used:

• Flask for backend API routing
• Perplexity Sonar to gather product information.
• YouTube's and Reddit's API to get comments, likes and views.
• VADER for sentiment analysis with a custom 5-point scale
• Custom-built scoring algorithms to evaluate relevancy and emotion
• React JS and charting libraries to power an interactive frontend experience

The project is structured with a clear backend/frontend separation and a modular Flask blueprint architecture.

Challenges we ran into

• Scraping reliable, high-volume data from social platforms within rate limits
• Calibrating a sentiment scoring system to reflect real user emotions over time
• Balancing a dual-metric model (relevancy vs sentiment) into a coherent UX
• Debugging Flask blueprint registration and API routing in a modular setup

Accomplishments that we're proud of

• Successfully built a fully functional sentiment/relevancy engine in under one week
• Created a visual interface that simplifies complex data for any user
• Built a pipeline that handles 5-year sentiment aggregation and real-time trend detection

What we learned

• The power of combining long-term and short-term metrics for deeper market insights
• How to orchestrate multiple APIs (Reddit, YouTube, Perplexity) under one architecture
• How vital clean UI/UX is in communicating technical data to non-technical users
• That team collaboration and quick pivots are essential in high-pressure hackathons

What's next for Goodsline

• Expand to TikTok and X (formerly Twitter) data sources
• Integrate multilingual sentiment analysis
• Add predictive trend modeling using historical sentiment + engagement patterns
• Package as a SaaS API for brands to plug into their analytics stack

Built With

• flask
• python
• react
• sonar
• vader