Arnold Kling rightly says that economic history can be used to teach macroeconomics more effectively. GDP presents a historical narrative of the evolution of GDP, which started as a result of war, and soon became the main macroeconomic indicator. So, yes, with a few explanations to clarify concepts, one can use the book in an introductory class on macroeconomics.
The Gross Domestic Product has been criticised frequently. Critics argue, among other things, that GDP is not a good measure of social welfare, that it does not include the informal economy, and that it does not deal well with innovation. Those criticisms are justifiable and spot-on, but as Diane Coyle argues, in spite of its many handicaps the GDP is still a useful tool to measure economic performance. From the beginning GDP was not seen as a measure of social welfare, but of income and production. Of course, we need other indicators to measure welfare, but as far as income and production goes, the GDP is still useful. That does not mean that it is perfect, not at all. In fact, it needs serious modifications.
I like Diane's writing style a lot. I follow her blog closely. However, I can not tell exactly what is so attractive about her writing. She explains her ideas and arguments with an almost optimal level of complexity, no more, no less. You can look at the way she describes GNP in the book. She could have given a long explanation but instead she just tells what is sufficient for understanding the main ideas in the book. The most important characteristic, I think, is that she asks big questions. At the end of the book she argues that, under the current digital and online economy, and everything that they imply we need a new definition of "the economy." That is a metaphysical claim of a mainly "physical" social science. So she makes one think big and not many writers can do that.
Of course I am making huge simplifications of sharply elaborated arguments.
Tyler Cowen said nice things about the book.