Predictor of successful retail stores: A small retail property recently became available about 2 blocks from me in NYC and, for fun, I began thinking about wha

Predictor of successful retail stores: A small retail property recently became available about 2 blocks from me in NYC and, for fun, I began thinking about what kind of stores would do well. I'd come up with an idea and realize that a similar type of store was just a block or two away. When I thought I'd come up with a good idea, such as a pet grooming service, I asked the nearby pet store what the traffic was like and apparently it was horrible. Enter predictive analytics. This tool would draw upon publicly available city records of what types of retail stores have been in neighborhoods that I can characterize through publicly available census data. If I could use the length of existence of a rented (not owned) retail shop as a proxy, I could, with X% probability, be able to predict, with probably surprising specificity, what kind of retail store would do well 2 blocks from me. This combines elements of what sort of retail store isn't serving that particular area (as an assumed factor of success), what kind of stores have failed in the past in the immediate area, which have been 'successful,' behavioral patterns of the neighborhood demographic (heck, combine that with social traffic tracked data), etc. Obviously, this discounts the innovation of retailers that adds to a store's success, among other factors, but this tool would at least yield an important factor to consider when trying to figure out what store would do well in a particular spot,