One of the things I have learned over my career is that I am very bad at predicting how much attention something I write will get.
Among my academic writings, my most cited paper is an article coauthored with David Romer and David Weil on the empirics of economic growth. It has more than six times as many cites in google-scholar as my second most cited paper, an article coauthored with Ricardo Reis proposing the sticky-information Phillips curve. When I was writing the paper with Romer and Weil, I had no idea it would get so much attention.
Similarly, when I write a column for the New York Times, I am not good at predicting how much it will get people talking. As a result, I monitor the subsequent blogosphere commentary to judge if the article is a snore (like most things that get published) or if it is commanding attention. At the very least, I expect my articles to be noteworthy enough that within a few days Brad DeLong will call me a moronic hypocrite. I hope my articles incite some wider commentary as well, but I never know in advance.
A case in point. According to the Times ranking, my most recent column is the 4th most blogged-about article that the Times has published in the past week. I did not expect such as reaction, as the point of the column--an explanation of Republican economic philosophy--did not strike me as particularly novel or controversial. The lesson, I suppose, is to keep writing what I think needs to be said, without trying to forecast public reaction.