Source Data

Below is an overview of our source data for Prospecting plan, some of which is also available in our open-data Search site (opens in a new tab).

Our Data Mission

When it comes to data, our goal is to give you the most fresh, fast, and free nonprofit dataset on the Internet.

Where Our Data Comes From

Our core dataset is taken from the raw IRS filings as published on irs.gov (opens in a new tab). If you are interested, you can access and download the same files yourself here - IRS Bulk Data Downloads (opens in a new tab).

Note: if you find yourself doing this for fun on the weekends, please call us to discuss coming to work with us at Grantseeker!

In our industry, this is the "cornerstone" of all nonprofit data. Every organization - Pro Publica, Candid/Guidestar, and many of our competitors - all pull from this same dataset. It's not always pleasant, but totally doable by most determined computer people.

Why the Data Isn't Always Correct

Yes. There are errors, mistakes, missing items in the dataset - look close enough at any one and you'll find something! Generally, there are three categories of error:

• Organization Errors: the organization submitted incorrect or incomplete data, either by filling out their 990 form or because they didn't have it right in the first place. This is not unusual; especially with newer filers.
• IRS Errors: IRS can make mistakes too! There are occaisionally transcription bugs, and even typos -- which you start to see when you're pushing millions of records into a database. Overall, they do pretty good work - and we should thank them for it! But mistakes can happen.
• Grantseeker Errors: We make mistakes too! The most common type of error on our account is one where we are "filling in the gaps" of missing data (e.g. for grantees - where the EIN is not provided). Sometimes misspellings or other glitches throw our systems off. For example, if a Funder lists a grantee as "GREATWORKS, INC." and another funder lists the same organization as "GREAT WRKS" (ie spaced and missing the "O"), our automated system might code it differently.

Despite these formidable data-munging challenges, we have one of the best datasets available -- and nobody makes more of it freely available than we do!

How to Correct Your Data

If you see something, say something! We take any and all corrections - just email "support at grantseeker dot io" and include in the subject line the magicwords "RE: DATA CORRECTION" so that we can flag it for review immediately.

Thank you & happy seeking!