Spilk: Chat Analyzer & Trends

Unlock the hidden stories in your messenger chats. Discover who texts more, texting patterns, ghosting habits, and more - all with on-device privacy at the core.

Comment

Inspiration

This project started from one evening of casual chatting. We were sharing stories about our dating lives, and having data science skills, it soon led to some non-trivial analysis of chats with our exes. That's how we came up with the idea for Spilk. We realized that the insights were interesting, amusing, and fun, and could be valuable to others too.

What it does

Spilk is a tool that gives people an opportunity to understand this aspect of their dating life better. It analyzes chat data from WhatsApp, Messenger, Instagram, and Telegram, providing insights about different conversation patterns. The app processes metrics like who texts more, who replies faster, who initiates conversations and more, presenting this information in digestible, shareable formats.

How we built it

Our design process is best characterized as unlearning. We started with complex charts that we personally found useful. But user tests quickly showed these were overwhelming. We then pivoted to reverse-engineering: mocking up digestible, shareable insights, figuring out what resonates with people who helped us test the app, and then we built it.

To ensure privacy, we implemented all analysis on-device, meaning personal conversation data doesn't leave the user's phone.

Challenges we ran into

The main challenge was making data analysis accessible to the average person without oversimplifying it. We faced several technical hurdles, particularly in processing diverse messaging formats. User privacy was a priority, but also a challenge. We had to refine the app without seeing actual conversation data, significantly constraining the feedback loop.

But despite the technical challenges, the biggest challenge was psychological: learning to see data through the eyes of potential users, not our own.

Accomplishments that we're proud of

We built an app that makes complex data analysis accessible and engaging for non-experts, working under significant privacy constraints. We condensed years of our professional and personal experience into something that can benefit everyone.

What we learned

We learned the importance of user perspective in design and tearing down many of our own assumptions.

What's next for Spilk: Chat Analyzer & Trends

We plan to refine our insights based on user feedback, potentially expand to other messaging platforms, and explore additional features that can help users improve their communication skills. Release an Android version of the app, and iterate on monetisation strategy. We hope our app continues to spark conversations, foster self-awareness, and help users find that perfect match.

Share this project:

Updates