When moving from the cloistered world of research to industrial use, some hard economic truths come into play. Often they can be the reason why a good idea fails to thrive, despite an apparently similar approach being hugely successful elsewhere. We will draw on decades of experience with domain-specific modeling in industry to show what can go wrong, why, and how to correct it. With many possible reasons, we will focus on those that are encountered time and again, have the largest effect, and differ most from what the audience might expect.