[vitd]

Very neat idea, and as others have said, cool animations! However, I'm not sure I buy the premise that "Social media signals can be used to give a rough picture of foot traffic around places of business," at least in general.

I can believe it for restaurants, hotels and popular hang-outs. But thinking about churches, strip clubs, the urologist, etc., I wonder if there are "blind spots" or other areas of low-quality data. Or even areas of poor cell coverage?

Maybe in most cities there are enough other places around that people frequent to cover over any blind spots? I'd be interested to know whether they've found any places like that?

[shayanjm]

Absolutely - this wasn't meant to be a be-all-end-all to lead generation by geographical bounds. This was more of a "I had a cool idea, wanted to test it, and here's how I went about doing that" sort of post. There are a ton of things that don't get picked up by social signals, but it's a low enough hanging fruit that I figured it was worth looking into to see if it was even a viable source of data at scale.

I think a 'real' analysis of the data at a much larger scale (probably state-wide would be a good test) would yield some interesting data on the effectiveness of this implementation in specific.

As mentioned in the post - the circle packing implementation itself is general enough to be used in other applications. The point was to show off a cool potential application for a solution to a non-intuitive problem.