Inspiration
We noticed that most trend tools tell you what already blew up, but not where something might blow up next. We wanted to make something actually useful for brands or creators trying to catch a trend early instead of being late to it.
What it does
ViralWave looks at YouTube trending data across 11 countries and helps show what content is performing, which countries have similar trend patterns, and where a content idea might have the best chance of taking off first.
How we built it
We built it as a Streamlit app in Python using Pandas, NumPy, and Plotly. We used the dataset to calculate virality scores, compare overlap between countries, and build features like the dashboard, virality predictor, and cross-cultural arbitrage page.
Challenges we ran into
The biggest issue was that our dataset was only a single snapshot, so we could not honestly predict exact timing of when a trend would move from one country to another. We had to rethink the product a bit and focus on cultural overlap instead. We also had to make sure the demo would not break if data or AI features failed.
Accomplishments that we're proud of
We are proud that we made something that feels like an actual product instead of just a chart or model. The cross-cultural arbitrage feature is probably what we are most proud of because it gives the project a more original angle.
What we learned
We learned that having data is not enough by itself. You still need to turn it into something that answers a real question for a real user. We also learned that being honest about what the data can and cannot do makes the project stronger.
What's next for ViralWave
Next we would want to use multi-day data so we could track how trends move over time, not just which countries overlap. That would make the product a lot better because then it could help predict not just where something might spread, but when.
Log in or sign up for Devpost to join the conversation.