On June 30th, the Chronicle of Philanthropy reported that President Obama announced

that White House officials will travel across the country to find “the most promising nonprofits in America” as the administration decides how to spend a new $50-million fund to help charities expand innovative social projects.

Of course the President's statement begs the question, what does "most promising" mean? The White House has spoken in vagaries about program effectiveness and scalability, but Betsy Fuch of the Modern Giving blog rightly points out that

With countless private foundations around the country squabbling to define the ideal due-diligence process, how does Washington know exactly how to build one? How will we be certain that those deemed the ‘best’ programs really are the best programs?

The real point here is that we simply do not know how to identify the "best programs" because we don't know what "best" means. The private sector enjoys a standard quantifiable metric of profit that allows businesses within and across sectors to be compared to one another. Social service evaluation is infinitely more complex.
Harvard business professor Clayton Christensen argues that the Obama administration should use

the convening power of the White House to initiate a focus on impact and metrics. Specifically, the White House should help initiate a process by which categories of social innovations are agreed upon, and metrics can be defined for assessing the impact of innovations in each category on the social problems that they target. Just like independent rating agencies have developed methods for assessing the safety of investments in various securities, methods might emerge that help social investors categorize the type of impact that various social entrepreneurs hope to achieve, and to rate the present and potential effectiveness of their efforts to achieve that impact.

While I agree with Professor Christensen that the White House should use its clout to move the industry toward evaluative metrics, I do not think the government should determine what those metrics are.
As Professor Christensen himself points out, in the private sector we rely on independent rating agencies who compete with one another do develop ever better evaluative metrics. We need a similar approach in social services, whereby multiple well funded, capable entities work on the serious problem of developing competing evaluative frameworks.
Perhaps the Obama administration's $50 million would be better used as research grants to develop a robust evaluation sub-industry in the social service sector. As it stands, without a clear definition of what "most promising" means, it is impossible to tell which, if any, organizations in the US are worthy of being funded by the Social Innovation Fund.

(Photo by wonderwebby)