Donor Profile Schema
Understanding donor profiles, confidence scores, and data sources
Every donor profile returned by the DonorAtlas API contains rich biographical, financial, and philanthropic data assembled from multiple public sources. This page explains the key concepts you’ll encounter when working with donor data.
DonorAtlas aggregates data from public records — nonprofit tax filings (IRS 990s), political donation records, corporate filings, news articles, and other web sources. Our identity resolution system matches records across these sources to build a unified profile for each individual.
Because these records come from independent sources that don’t share unique identifiers, DonorAtlas uses statistical matching to link them. This is why certain fields include a probability score.
The full donor profile (APIDonor) contains the following fields:
Several fields on the donor profile include a probability value between 0 and 1. This represents DonorAtlas’s confidence that the record belongs to this specific donor, not the probability that the event occurred.
For example, a board_affiliation with probability: 0.92 means DonorAtlas is 92% confident that this board membership record belongs to the donor shown in the profile — and 8% that it belongs to a different person with a similar name. The record itself is from an authoritative source (typically an IRS 990 filing).
You can filter results by probability in your application to control data quality. For most use cases, a threshold of 0.70 or higher is recommended.
Many sub-models (Education, Work, BoardAffiliation, Donation, PrivateFoundation) include a source_urls array containing URLs where the information was found. These are the original public sources DonorAtlas used to build the record — typically nonprofit websites, government databases, news articles, or corporate filings.
The top-level source_urls field on the donor profile lists all source URLs aggregated across the entire profile.
net_worth_min and net_worth_max are the estimated bounds of the donor’s net worth in dollars. These are derived from real estate records, political giving patterns, nonprofit giving patterns, and other public financial indicators. They are estimates, not exact figures.
The net worth estimate includes explanations of the contributing factors via nonprofit_stats.capacity.explanations. Each explanation has a title (e.g., “Real Estate Holdings”) and a reason describing how that factor contributes to the estimate.
political_stats.capacity is the predicted political giving capacity for the current 2-year cycle.
nonprofit_stats.capacity is a NonprofitGivingCapacity object containing:
predicted_annual_giving — the predicted annual nonprofit giving capacity in dollarsexplanations — a list of factors that contributed to the estimate, each with a title and reasonBoth are forward-looking predictions based on historical patterns, not historical totals.
donations[].amount_range is a [min, max] array representing the estimated total amount donated to a specific nonprofit. Because nonprofit donation records are most often reported in bands (e.g., “499”) rather than exact amounts, the API returns a range.
nonprofit_stats.total_amt_range follows the same pattern for the aggregate total across all nonprofits.
top_issues is a list of philanthropic interest areas (e.g., “education”, “environment”, “healthcare”) that DonorAtlas has inferred from the donor’s giving history, board affiliations, and biographical information. Each issue includes a description explaining why it was assigned.
is_deceased indicates whether DonorAtlas has found evidence that the donor is deceased. A value of null means the status is unknown — it does not necessarily mean the donor is alive.
date_last_updated indicates when DonorAtlas last refreshed the donor’s profile. Profiles are updated periodically as new public records become available. You can use this field to assess data recency.