Every year, Candid processes data on nearly two million organizations and more than three million grants. That data makes its way into various products and services to help: nonprofits find funding; researchers, advocates, and journalists derive insights into what is happening in the sector; and all types of funders to make funding decisions. All this work would be significantly more difficult if it weren’t for Candid’s taxonomy, the Philanthropy Classification System (or PCS). A taxonomy is simply a system of classification, or a way of organizing things. We’ve all encountered taxonomies, like the more formal biology classification system we learn in middle school (remember kingdom, phylum, family, genus, species?), or the style, size, color, and other categories we use to find what we’re looking for when online shopping. In Candid’s case, the PCS is one of the key things that makes our data usable. Without it, people would be left to dig through a mountain of data to try to find the information that’s relevant to them.
This year, Candid is updating its taxonomy. Based largely on input we’ve received from individual organizations and partners, we’re proposing 205 changes to the PCS. Before moving forward with these changes, however, we want to give as many people as possible a chance to weigh in on them. Starting today and through August 5, 2023, we are making the proposed changes available for public comment. The PCS is meant to reflect the work of the sector; it’s important to us that the sector has a chance to shape it.
How we got here
The Philanthropy Classification System was developed in 2015 by what was then the Foundation Center (one of Candid’s predecessor organizations), which found that the National Taxonomy of Exempt Entities (NTEE) used by the Internal Revenue Service (IRS) did not adequately capture the work of the sector. The PCS expanded on the NTEE and covers five facets: subjects; populations; support strategies (e.g., program support, general support, advocacy); transaction type (e.g., cash grants, program-related investments); and organization type. Candid codes much of its data according to the PCS, including organizations, grants awarded, funding opportunities, news, and more.
In early 2022, Candid solicited recommendations on how to improve the PCS. This is only the second time Candid has sought to update this taxonomy, and the first time since the PCS’ release that significant external input was solicited. Groups could recommend: adding, combining, or removing terms; moving terms in the hierarchy; or updating term definitions. In all, we received over 600 recommendations from nearly 30 organizations, as well as from Candid staff with extensive knowledge of the taxonomy.
These recommended changes then went through multiple rounds of internal review. During this review process, which took longer than expected due to the number and complexity of the recommendations we received, staff considered whether:
- A proposed term was already covered by an existing PCS code (if so, we rejected the proposed change)
- A proposed term could be applied to a sufficient number of grants or organizations (this is to avoid adding terms so narrow that few search results are returned)
- A proposed term would be intuitive to users of Candid’s products, as well as a more general audience
- A proposed term is widely used or aligned with standards set by more recognized organizations, such as the Associated Press (which maintains a stylebook)
- A proposed term was actually in use in our existing data, such as in mission statements or grant descriptions (since the PCS is meant to describe the work of the sector as a whole, our goal here is to avoid terms used by only a few organizations)
- It would be feasible to apply the proposed term or make the proposed change based on the quality of the data available
- There was a compelling rationale for combining or moving terms within the hierarchy, since these types of changes in particular can be disruptive
- The proposed change would negatively affect the overall coherence and balance of the taxonomy
Once we had applied these criteria, we were left with 205 proposed changes, broken down by facet and change type in the chart below.
| Facet | Change definition |
Change term label |
Move term |
Combined terms |
Remove term |
Add term |
Total |
|---|---|---|---|---|---|---|---|
| Subjects | 48 | 21 | 31 | 2 | 3 | 19 | 124 |
| Populations | 24 | 31 | 3 | 0 | 4 | 9 | 71 |
| Support strategies | 2 | 0 | 0 | 0 | 0 | 0 | 2 |
| Transaction type | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
| Organization type | 0 | 0 | 0 | 0 | 0 | 7 | 7 |
| Total | 74 | 52 | 34 | 2 | 7 | 36 | 205 |
| % of total | 36% | 25% | 17% | 1% | 3% | 18% | 100% |
The PCS will never be perfect–the sector and society itself change too quickly for that to be possible–but it can always be better. We’re sincerely grateful to the organizations and individuals who have made such thoughtful recommendations to date, and we look forward to receiving broader feedback during this public comment period. Candid staff will review these comments in August and September and share the finalized changes with the sector in the fall.
