Imagine if the maker of some new, high-budget PC game called "DRM GoreShooter" decides that they will be in good shape if they make $5 million in the first three weeks of sales and past that point they won't really be profitable anyway (DRM or not). They want a profit better strategy than trying to use DRM to extend the window of pre-piracy sales out to week four or five, which might or might not work anyway. So they sell the game with something physically in the package that's really cool and you can only get in the first few weeks and they let you report in your product code, one time, for credit on Facebook or wherever for you having sponsored the development of the next game and been a generally cool person paying for quality entertainment. Would that be a better approach? It might ultimately result in the same money, but could be done as a way to cost less or at least provide a great marketing push and build good will in the fan-base.
My argument would work better if I had any idea how much it costs to develop, maintain and manage a per-title DRM strategy, I suppose.