Theories of deliberative democracy, and democratic theories more generally, have suffered from a great deal of imprecision concerning what the relevant object of deliberation ought to be. To the extent that they have implied that deliberation should have to do with abstract, single-issue moral positions, they have erred in at least two ways. First, part of what democratic deliberation must deal with is trade-offs and compromises *between* positions on a range of policy issues, rather than on single issues. Second, deliberation should have to do with policy proposals, that combine an abstract position of policy principle with concrete considerations (among other things) of resource allocation and use of coercive state powers. I argue that policy platforms that are typically developed by political parties -- integrated statements of policy that combine policy proposals on the full range of issues that government has to address, and that make plain the means that will be employed in order to realize policy objectives -- satisfy these two desiderata, and are thus appropriate objects of democratic deliberation. The paper is divided into four sections. The first criticizes prominent theories of deliberative democracy and of "epistemic democracy" for not taking considerations of trade-off and implementation seriously enough. The following sections consider the moral issues that arise in trade-offs, and in the two main dimensions of policy implementation, namely resource allocation, and use of coercion or incentives by the State.