Saving is hard for most people (unless you’re Chinese or German it seems). A typical approach to encourage savings might be to emphasize all the good reasons to save (children’s education, healthcare needs, retirement, etc.) and set savings goals for these reasons.
This blog has now featured a healthy debate between researchers advocating randomized evaluations and those cautioning the overuse of such methods. One point that I believe both sides would agree on is that irrespective of which empirical methods we use, it is important to understand and analyze the causal chain of impact. Such analysis can greatly enhance the external validity of any evaluation.
There are now a variety of well-known experimental and non-experimental methods that economists use to learn whether a given program works or not. However, our tools for learning why or why not something works are much more limited.