It may seem unfair, but I think the reason I would give as to why it isn't is: you have to pay for that source of data. Using satellite imagery to analyze sales

It may seem unfair, but I think the reason I would give as to why it isn't is: you have to pay for that source of data. Using satellite imagery to analyze sales, imho, sort of falls into the same area where HFTs pay for feeds from exchanges so they can take advantages of widening gaps in bid/ask spreads. When it comes to insider trading, it really seems to boil down to: did I obtain the information in a way where no outsider would have possibly done so? With your example, it certainly seems possible that other groups could perform the same analysis. So as you said, it wouldn't be insider trading. I think another one of the aspects of insider trading is _certainty_. If you're an insider, and you know your company is going to miss earnings, the likelihood of you making a wrong bet are near 0. If you're analyzing imagery of parking lots to see if sales at Wal-Mart are going up, theres still a non-insignificant chance that you could bet wrong. Maybe the reason theres so many cars it that they are running promotions that actually don't increase revenues in an expected way. Maybe there aren't any cars because people are taking public transit. When you're an outsider, theres way more variables that go into your trade, compared to when you're an insider and its a simple question of "how will this information about my company get reacted to?" Insider trading is probably one of the most interesting aspects of modern finance. As you can probably tell, I take most of my information from Matt Levine (he gets linked here pretty often, but his column is Money Stuff on Bloomberg (also an email newsletter so you don't use up your free articles)) Again, not an expert, take everything I say with a healthy pinch of salt.