Inspiration
We were inspired by platforms with well-functioning tag systems that genuinely benefit both creators and audiences like Archive of Our Own, where tags aren't just metadata but a powerful discovery tool. We wanted to bring that same philosophy to YouTube: what if creators could use tags and keywords not just to label content, but to understand what's actually working in their space?
What it does
StatPatty is a keyword-based content intelligence tool that helps creators understand what works and what doesn't, using real YouTube trending data. Users enter a keyword and the system searches across video titles, tags, and descriptions to surface the highest-performing content in that niche. It ranks videos by views and engagement, identifies top-performing tags and categories, and flags underperforming or controversial content using like/view/comment ratios. Users can also filter out specific tags, creators, or topics to zero in on their exact niche and compare trends across different countries with one click.
How we built it
We built StatPatty using a real YouTube trending dataset as our analytical foundation, engineering key metrics like engagement rate, like ratio, and comment ratio. We used AI tools including Claude to help develop the website and accelerate the build, allowing us to focus more of our time on product design and feature logic rather than getting stuck on implementation details.
Challenges we ran into
We spent a significant amount of time in the ideation phase, debating features and scope before committing to a direction. This left us with less time to build than we'd have liked. It was a good reminder that a clear, scoped plan early on is just as important as the idea itself.
Accomplishments that we're proud of
We shipped a working product. Given the time constraints and the scope of what we were trying to build, having a functional tool that actually searches, filters, and surfaces insights from real data is something we're genuinely proud of and the core concept works.
What we learned
There are far more factors behind a trending video than we initially expected. View count alone tells you very little, engagement rate, like ratio, comment patterns, tag strategy, and even where a keyword appears (title vs. description) all tell completely different stories. Building this tool gave us a much deeper appreciation for how nuanced content performance really is.
What's next for StatPatty
We want to integrate live YouTube API data so the platform reflects real-time trends rather than a static dataset. We also plan to make an AI analyzer layer to generate personalized, actionable content briefs, so any creator can walk away with a concrete strategy grounded in what's actually working right now.
Log in or sign up for Devpost to join the conversation.