CARE is launching a new evidence strategy for a rapidly changing world. We aim to transform hundreds of millions of lives, even in a world of declining resources. To do that, we’re investing in real-time data tools to evaluate our work the same way disciplined product teams evaluate their investments: by cost, quality, user feedback, and potential for growth and scale.
We’ve restructured our entire research and evaluation function to consistently answer this strategic question: in this context, will this deliver more long-term impact per dollar than the next best option?
Our new evidence strategy combines rigorous proof, sharper, more frequent analytics, and a deep investment in cost effectiveness across different contexts. This becomes the core operating discipline for deciding what we design, fund, adapt, and scale, not a retrospective reporting exercise.
Our strategic discipline also listens to the communities who will live with the results and the local leaders and governments who roll out and sustain the solution. Those voices determine whether a validated model scales or fails to deliver durable results in the real world.
CARE is uniquely positioned to bridge proven solutions and the local ownership that’s required for true success. We’re harnessing our existing infrastructure—deep community trust, massive scale, rapid absorptive capacity, and our commitment to transparency and constant improvement. The local experts in our network understand the context, the systems, and the stakeholders that decide when a high-performing solution becomes a system-wide change that lasts.
Impact for the people we serve is our key goal. That doesn’t change. What has changed is an upgraded operating system built onthe tools we need for the next 10 years instead of retrofitting old systems for incrementally better results. Our model is built with credible research partners such as the Abdul Latif Jameel Poverty Action Lab (J-PAL), UC Berkeley’s Center for Effective Global Action (CEGA), Harvard Center for International Development, and IDinsight.