Imagine if we can codify the world’s evidence base and learn what works, and what doesn’t. We could have greater confidence in what works — by solving the reproducibility problem. The issue isn’t really about reproducing results — we only really want to do that so we can have confidence that a particular solution will work more than once.

But we can solve that problem in a different way — instead of trying to prove that a program worked and can work again — we can instead tease out and understand the underlying DNA of the program, and program’s like it. In this way we’re not reproducing a program, we’re isolating the variables that drive program results so that we can reproduce those variables.

Empowering all practitioners to use evaluation tools and access evidence, rather than hiring consultants or waiting for experts to tell them if they’re effective and leveling the Playing Field so all programs could be evaluated on equal footing.

Not just because they’re big or were funded in the past.

Helping funders understand which programs produce the outcomes they desire, and at the same time, helping practitioners clarify the total cost of ownership of an outcome, i.e. what it really costs to produce results.

Lowering fundraising costs and reducing yield loss for funders and investors who all too often don’t get the results they want, which limits future investment.

By defining programs in a binary way, this works, this doesn’t, and by analyzing programs as a black box rather than understanding their underlying structure and design, we are limiting the capabilities of using past evidence.

And finally, the current framework of evidence-based thinking can be stifling to innovation. Implying that a program can’t work because it hasn’t worked before, and limiting choices of funders or policymakers only to proven models is highly problematic.

We want to find ways to use knowledge and evidence to inform innovation, not stifle it.