ReFED Insights Engine.

Data and solutions for food waste.

ReFED is a national nonprofit dedicated to ending food loss and waste across the U.S. food system by advancing data-driven solutions. Together we created a platform to achieve exactly that.

Collaborating with key audiences to generate large-scale, meaningful impact.

38% of all food in the USA is wasted.

The equivalent of 149 billion meals annually. It's valued at $444 billion, roughly 2% of U.S. GDP, and it has the same climate footprint as the entire U.S. aviation industry (passenger, commercial, and military).

Data-backed solutions.

The platform provides an extensive cost-benefit analysis of 40+ solutions.

Centralizing knowledge.

Built with data from more than 50 public and proprietary datasets.

A worthy investment.

10 years of $14b annually could reduce food waste by 45Mt and a benefit of $73b each year.

Developing the ReFED platform.

The brief.

From the beginning, ReFED sought to bring new philanthropic and investment capital, along with technology, business, and policy innovation, to the food waste challenge. Aiming to achieve a 50% food waste reduction by 2030, the technological arm is an essential part of pooling all the knowledge together to inform the necessary action.

Our role.

We worked with them to develop the ReFED Insights Engine — a tool that provides information, insights, and solutions to anyone interested in food waste reduction. Our role was to build the front-end analysis tool and develop a back-end data storage system capable of handling more than 3 million facts, from 50 different public and proprietary datasets. The result is a tool that food businesses, investors, innovators, nonprofits, policymakers, and more can use to solve food waste problems.

Delivered with impact.

Lots of data doesn’t have to mean slow.

Creating a frontend as easy to use as the ReFED Insights Engine requires extra care in the backend. The key to success is a good definition of the filters and aggregates that the system needs to offer. Once those were defined for the Food Waste Monitor, we optimized the database for specific types of queries using a data analysis solution that reduces the original facts table from more than 3 million rows to less than 40 thousand. As a result, a search query can return data in around 150 milliseconds.

Mongabay Data Journalism Tools.

Environmental data journalism.

Want to make a difference together?

Let's talk