Anne Sprinkel and Dipendra Sharma from CARE's Tipping Point project talk about the challenges in implementing RCTs, and the risk of sacrificing communities' needs to the methodological rigor that researchers demand. "Make sure you have a good reason for doing an RCT," says Sprinkel. Sharma adds, "Start with good programming, then build research around it." They also have some great tips for managing expectations, clear communication, and just how long it takes to do it right (Hint: it's a lot longer than you think).