Whether you’re a nonprofit leader, a funder, or an analyst in the social sector, you’ve probably spent hours poring over funding data, trying to better understand what the funding landscape looks like for your specific cause. Trust me, I feel you.
Candid’s data is widely used to understand grantmaking in the sector, and for good reason: Our data is detailed, comprehensive, and searchable. But these virtues can be a double-edged sword. In fact, because Candid’s data collection efforts have expanded and diversified in recent years, sometimes data needs to be removed to provide the clearest picture of funding on a specific topic.
From one nonprofit data nerd to another, here are three pitfalls to avoid—to make sure you’re not artificially inflating how much money is going to your cause.
1. Don’t count on pledges
“Pledges” reflect a funders’ public promise to dedicate a certain amount of funding or resources to address a crisis or issue, often over a number of years. Pledges can be an important early signal of funding trends and therefore provide insight into future giving. In 2020, Candid increased our efforts to include pledges in our databases to capture emergent trends in giving during a year of crisis.
However, pledges should not be interpreted as money in the bank, as there is no contractual obligation to fulfill these promises. It’s also important to know that pledges always remain categorized as pledges in Candid’s database. This is because it’s virtually impossible to tell whether a funder has made good on a given pledge. Pledges tend to be broad and vague (e.g., “Funder X pledges $5 million to support education over the next decade”), while grants tend to be narrow and specific (e.g., “Funder X awards $50,000 to Nonprofit Y in 2022 to conduct a research study on standardized math tests in public high schools”). This mismatch makes systematic deduplication efforts impossible to do reliably and accurately. At the same time, adding up pledges and other transaction types risks double counting, as some grants may in fact be pledges fulfilled.
When conducting our own analyses, Candid’s research team always treats “pledges” separately (either by analyzing pledges separately or removing them from the analysis), and we recommend that you do as well. We also often filter specifically for “cash grants.” Cash grants represent 99% of Candid’s data and are distinct from other grantmaker types in-kind gifts, pro bono services, and program-related investments.
2. Don’t forget to filter by funder type
Candid is perhaps best known for our detailed and comprehensive collection of U.S. foundation grants. However, that’s not the only type of grantmaking data we collect. Over the last decade or so, we’ve been expanding our data collection efforts to include grantmaking from public charities, corporations, non-U.S. institutions, and occasionally government spending and donations from high-net-worth individuals (e.g., MacKenzie Scott).
Totaling dollar figures across different funder types can muddy the waters when we try to understand funding patterns, especially as the completeness of Candid’s data varies greatly by funder type. Candid will eventually have a comprehensive database of funding for organizations that file IRS Forms 990: U.S. community foundations, independent foundations, public charities, company-sponsored foundations, and operating foundations. However, other funder types are not required to disclose grantmaking to the IRS; therefore, information is limited to what they choose to share publicly (through press releases, websites, etc.).
If you’re trying to understand U.S. foundation grantmaking specifically, the Candid research team recommends removing government/federal funders, individual donors, and public charities from the search criteria to get a true picture of giving by all private grantmaking institutions—which makes up 58% of Candid’s database of recent grantmakers. Or, if you’re interested in grantmaking from independent foundations only, you can filter by that grantmaker type, which accounts for about 44% of Candid’s database of recent grantmakers. You’ll also want to limit your search to grantmakers based in the United States.
3. Don’t count regranted dollars twice
Another case where double counting is possible is when grants are awarded to other funders for regranting purposes. For example, a private funder may donate to community foundation, who would then “regrant” that money to a nonprofit such as a local food pantry. This type of funding flow can get tricky if you’re trying to calculate exact dollar amounts granted to a specific cause, because if you’re not careful, you might count the same dollars twice.
The Candid research team accounts for regranted funds by flagging and filtering out grants that have been awarded to organizations that also appear as grantmakers in the same data set, as that often indicates that the same dollars are being regranted and probably should only be counted once.
When in doubt, be a data detective
Data about the social sector is powerful, but it’s also messy. On the Candid research team we often joke that our true jobs are being “data detectives” following hunches, running down leads, and conducting investigations to get to the true nature of the situation. I recommend you do the same. A little curiosity, cleaning, and filtering can make all the difference when it comes to understanding grantmaking data. To learn more about Candid’s grants data, check out our research manual.
The post 3 pitfalls to avoid when analyzing the dollars going to your cause appeared first on Candid insights.
