I started this because I kept seeing headlines making bold claims about cities, like "gentrification is displacing X" or "income predicts Y," but they rarely showed the underlying data at a granular level. I wanted to be able to look at a city and actually see the patterns for myself, ZIP by ZIP, rather than taking someone's word for it. The original idea was simple: put Census data on a map and let people compare metrics. But once I started pulling in the data, I realized how fragmented and inconsistent city-level open data really is. Some cities publish crime data with lat/lng coordinates in clean CSVs. Others bury it behind PowerBI dashboards or require FOIA requests. That friction became part of the project itself, figuring out how to normalize 16 different cities worth of data into something comparable. The scope kept growing naturally. I'd add income data, then wonder how it correlated with health outcomes, so I'd pull in CDC data. Then I'd wonder if chain store density tracked with income, so I'd pull in OpenStreetMap data. Each new layer raised new questions, which is kind of the whole point of the tool. The bivariate choropleth was a turning point. Once I could overlay two metrics on the same map, it stopped feeling like a data viewer and started feeling like an analysis tool. The correlation engine followed from there. At some point it shifted from "I want to see this data" to "I want anyone to be able to explore these questions without needing to know SQL or Python."

Correl Data

Explore correlations in places and people across US cities

Explore correlations in places and people across US cities