Start with strategy.

If you have a science that cannot afford to be misread, tell us what it has to do and who has to believe it, and we will tell you how we would approach it. Three ways in.