Categorization can help us make sense of the world: A family sorts its mail, an investor allocates among asset classes, and a child organizes her candy after Halloween.

In “Pay-What-It-Takes Philanthropy,” Bridgespan’s analysts make a compelling case that before we can meaningfully benchmark nonprofits’ program and indirect costs we need to sort organizations into segments. The nonprofit sector is too big and too complex for us to compare all nonprofits in a single group.

The “overhead myth”—the misguided idea that nonprofits’ performance can be judged by accounting ratios—has many flaws. Perhaps the biggest flaw is this: Nonprofits differ wildly from each other. It is incoherent to compare financial ratios between a university with a billion-dollar endowment and a homeless shelter.

This article offers a new way of thinking about comparisons among nonprofits. The four categories it offers are only a beginning. The authors rightly call for more research to build out a robust taxonomy of nonprofit business models.

The timing is good. The nonprofit field is rethinking the ways it categorizes itself. Existing frameworks—for example, the National Taxonomy of Exempt Entities—have provided an important foundation, but we must work to reach the next generation of nonprofit categorization. The Foundation Center has offered the Philanthropy Classification System as a new, multi-axis kind of taxonomy for philanthropy. Recent work by GuideStar to create a categorization of nonprofit performance metrics (the Common Results Catalog) offers another layer of richness to how we understand the actual work of nonprofits.  

Our challenge as a field is to connect the kind of financial categorization offered in this article with our growing store of programmatic and operational information. With work, we will be able to present nonprofits as more than lists of EIN (employer identification numbers) numbers and context-less financials. Instead, we can explore organizations within natural groupings such as art museums or environmental advocacy groups. (The right taxonomy need not be a perfect taxonomy—it is okay to have a big group for “other”!)

Many of GuideStar’s users and partners have called for a simple star-rating system for nonprofits. We resist those calls. As a field we must avoid oversimplification. Let us talk about nonprofits in a multidimensional way—and within appropriate categorizations. Nonprofits are complex entities and deserve to be understood as such. Humanity has proven itself capable of making complex, multidimensional decisions in other arenas of life: We do it all the time when buying a house or planning a vacation. Let us do the same when it comes to nonprofits.

To do so, we need help. Nonprofits need to be willing to share the data necessary to tell a multi-dimensional story. At GuideStar, we gather and disseminate information about every single IRS-registered nonprofit organization using the 990 form as our foundational data source. But we also ask nonprofits to voluntarily share additional data about programs, operations, and finances by updating their own profiles. To date, 115,000 have given us some additional data, with 37,000 providing enough data to earn one of our transparency seals. In May we launched the newest seal level, Platinum, for organizations that share quantitative programmatic data. Just like indirect costs, programmatic data is most meaningful when we categorize organizations not by their tax status, but by the nature of the work they do.

This is no small task. What we’re talking about will take a decade or longer. But as this article shows, we are making progress in how we understand the complex work of social change. Natural categories that reflect the diversity of the nonprofit sector can help us offer meaningful, contextualized data. And that can drive more giving, smarter giving, and a more effective field.

Read the rest of the responses.