I would like to add that one of the points in the discussion was a simple application of a much broader point. Chen says, "It's bad enough that your own marketi

I would like to add that one of the points in the discussion was a simple application of a much broader point. Chen says, "It's bad enough that your own marketing efforts drive down channel performance, but usually once your marketing efforts are working, your competitors quickly follow." The broader point is to think about the world in terms of feedback loops -- what we might call feedback theories. Our modern understanding of nuclear chain reactions works this way; once you get above a certain density, a neutron will be likely enough to hit a fissionable isotope, that the average number of neutrons produced per neutron is more than 1.0 -- then the neutron makes itself more likely, the chain reaction grows exponentially with positive feedback, and boom. Otherwise it will just peter out to 0 from negative feedback. Evolution by natural selection is also a feedback theory, if you focus on the genes themselves. Coming back to the economic examples, it is now considered a relatively good economic model to assume that the stock market already knows everything that there is to potentially know about a company. If you just imagine that there was some person who magically started moving money to the day's top performers, you can reasonably envision that others might try to jump on this magical predictor, then attempted to bail out just before they did -- but so many would potentially sign up for this that it would damp that predictor's own profits to have a massive sell-off just before they were supposed to sell themselves, negative feedback. It's thought that on a smaller scale this would tend to happen with insider knowledge. Or, for another econ 301 example, there is a powerful argument that competitive markets are efficient, because they must be among the least profitable markets around. Why must they be the least profitable? Because if competing is easy, then anyone in a less profitable market cuts their losses and jumps into the competitive market instead, which concentrates profits in the market they're leaving and dilutes profits in the market they're joining. And if there's no profit, there's no room for inefficiency. (There's also a sort of odd pseudo-Marxist corollary: that agents within a market always want the market to be less competitive, so that their profits can be higher.) Anyway, the law of shitty clickthroughs amounts, I think, to a particular feedback theory of marketing. If you market well, that generates negative feedback, both in the form of people following your example, and in the form of people getting desensitized to your ads.