"Instacart CEO Apoorva Mehta told AllThingsD that his company built a 'feasibility matrix' to determine which city it would expand to. Components of the matrix included what you might guess: car-ownership density and income. Weather was reportedly also a factor. Mehta tells TechCrunch that '33 percent of the year, Chicago has some form of precipitation' and that his company has learned in the San Francisco area that 'people just do not like grocery shopping when it’s raining.' Seattle, you might be next."

I think Chicago is a natural fit. The specific neighborhoods they picked have about 600k people in 33 square miles (a bit denser overall than San Francisco as a whole). There are a ton of residential skyscrapers in those neighborhoods, which means they probably can hit several orders in one go. 30% of Chicago households don't have a car, about the same as San Francisco. It rains or is cold about half the year, and is hot and humid a couple of months out of the remainder, which makes people not want to carry groceries home.

Agreed. Being from Chicago I think that its very natural for all the points listed above. The density of the city with the void of competition in the space (only Peapod comes to mind) make Chicago a great market to expand too. Chicago usually doesn't come to mind first as an early adopter market but given the rapid adoption of Uber (like other cities) I think the market is proving it is interested in similar services.

Agreed. I know many people who don't have a car living in the denser parts of the city, and grocery shopping (even for a single person living alone) can become a harder task simply because you can't carry a ton at once.

Yeah, the third largest city in the US with 2.7 million people. Perhaps you've heard of it?