Collecting data during a humanitarian crisis

When a crisis occurs—whether it’s the war in Ukraine or the COVID-19 pandemic—people seek up-to-date information. They want to understand, for example, how the crisis is evolving or how many people are affected. In the social sector, the questions that often emerge are: Which organizations are addressing the crisis? Who is funding them, and what activities do donors support? As the largest source of data on philanthropic and nonprofit organizations, Candid fields many of these questions.  

With breaking news, it takes time for the full picture to emerge. Weeks and months can go by before we understand what’s really happened, or why. Candid faces similar challenges in answering questions about the philanthropic response to current events—but with an even more significant time lag. It takes more than two years to collect comprehensive data about U.S. foundations’ giving. [1] In the midst of a crisis, waiting two years is simply not an option. To help the sector understand the emerging funding picture, Candid prioritizes collecting current, “real-time” data. This data is patchy, incomplete, and messy, but it’s often the best that’s available in the moment.

Here we share Candid’s processes, the data’s uses and limitations, and a call to action.  

Collecting and using “real-time” data during crises 

When a humanitarian crisis occurs, Candid mobilizes a team to collect information about the philanthropic response. We review the flurry of news stories, press releases, and announcements, and we actively search websites to identify and record grantmaking data. Candid also alerts the cohort of nearly 900 funders who share their grants data with us. We inform them that we are tracking a particular crisis and invite them to send us information about their grants related to the crisis. The data from these various sources are harmonized, and each record is coded according to Candid’s Philanthropy Classification System, either by Candid staff, our machine learning model, or a combination of both. 

This data is then made available in Candid’s Foundation Directory and Foundation Maps, as well as for use in research. The most recent examples of Candid research using real-time data are the COVID-19 report, produced in partnership with the Center for Disaster Philanthropy (CDP), and updates via our blog about philanthropic funding in response to the Ukraine crisis. We also made the data publicly available through special issue pages about the war in Ukraine and coronavirus, where users can access the data, alongside news stories and additional resources.  

Candid shares real-time data to provide those who want to help with the information they need to make better decisions. For grant seekers, the data shows who’s providing funds to address a crisis. Funders can learn how much their colleagues are giving and what they support, find opportunities for collaboration, and identify gaps.    

What to know about real-time data 

Unsurprisingly, the unique circumstances and methodology of this collection result in a unique dataset. Below, we outline some of the most important things to know about this data—especially for journalists, researchers, and others seeking to leverage real-time data to identify funding trends.  

• Candid’s real-time data is skewed toward large donors from English-speaking nations who publicize their giving. The data should not be assumed to be representative of all donors or all grants. While we recognize that a global crisis fuels a global response, as a U.S.-based organization, Candid’s data collection relies heavily on sources written in English. Real-time data also tends to reflect the largest contributions, primarily from institutions and high-net-worth individuals, as these are the types of donations that make headlines. Corporations, for example, are often among the first to announce funding for a crisis and, therefore, tend to be more represented. Many funders do not publicly announce their grants or share information with Candid, and millions of people give donations directly to nonprofits, to crisis funds, and through crowdfunding platforms. These gifts are not generally captured in our real-time data collection. Those who use this data for analyses and reporting should take these factors into account.  
• Real-time data includes a combination of grants and pledges. Grants are funds that have been disbursed or, at least, committed to specific organizations. Pledges, on the other hand, are announcements of intent to make a financial contribution—in other words, a public promise of future support. Candid does its best to parse through the messaging to understand if an announcement is a pledge or a grant, and we code it accordingly. Because of the challenges of aligning public promises with later grant dollars, Candid aggregates pledges and grants separately in our analyses. We recommend that others do the same. 
• Analysis that relies on real-time data does not provide a final, definitive picture of funding. As mentioned earlier, it takes Candid more than two years to collect comprehensive data about U.S. foundation grantmaking. Any analysis completed in the same year of a crisis (or even the year after) relies primarily on real-time data. The funding picture will shift, and analyses may look different from one moment to the next as Candid continues to gather additional data. Researchers, journalists, and readers should keep this in mind, as we collectively try to make sense of the latest trends in philanthropic giving. 

Call to action 

Despite its limitations, real-time data contains valuable early indicators of the philanthropic response to a crisis. The best way to enhance this data is for more funders to share information about their grantmaking. For institutions making public pledges of support, it is just as important to share where and how funding is ultimately directed. Greater transparency helps build public trust and advances a more efficient, effective, and equitable philanthropy. Check out this blog to learn more about why funders voluntarily share their data.  

[1] See Candid’s 2021 blog assessing delays in data collection from U.S. foundations. We will never have the complete picture of giving by corporations or individuals, since these donors are not required to publicly disclose details about their gifts.