How Tax Accountants Find Nonprofit Clients Using IRS 990 and SOS Data

Every CPA firm that targets the nonprofit sector faces the same prospecting problem: finding organizations that are large enough to need professional tax help, new enough that they haven't locked in a long-term engagement, and local enough to actually reach.

The data to answer all three questions is public. Most firms just don't know how to use it systematically.

Why Nonprofits Are a Good CPA Target

Nonprofits file Form 990 annually. That document is a treasure trove: total revenue, total expenses, executive compensation, program descriptions, board composition, and the name of the preparer who signed the return.

That last field is the one tax accountants care about. If a nonprofit lists a sole practitioner or no paid preparer at all, they're underserved. If their 990 shows revenue growing past $500K but they're still doing it in-house, that's a transition moment — exactly when a firm can add value.

The 990 also tells you whether the organization is financially stable (program expenses as a percent of total), whether compensation ratios look reasonable, and what their audit status is. All of this is screening information that would take hours to gather manually from individual organization websites.

The SOS Layer: Finding New Organizations Before They File Their First 990

Here's a timing advantage most accountants miss: 990 data lags by 12–18 months after the end of a fiscal year. By the time you see a new organization in the IRS database, they've probably already filed their first return — possibly with a competitor.

Secretary of State filings are real-time. When a nonprofit incorporates in California, the filing appears in the CA SOS database within days. You can monitor new nonprofit incorporations by entity type (usually "nonprofit corporation" or "benefit corporation"), filter by county for geographic targeting, and reach organizations before they've hired their first accountant.

The workflow:

1. Pull new nonprofit incorporations from state SOS databases weekly
2. Cross-reference against IRS 990 records to identify organizations that have been operating for 2–4 years (past startup chaos, not yet entrenched)
3. Filter by revenue band from 990 data ($250K–$2M is the sweet spot for CPA engagement)
4. Flag organizations with no paid preparer or a preparer who isn't a CPA firm

That's a targeted prospect list that would take a junior associate days to build manually. Automated, it runs in minutes.

Building the Data Pipeline

The two data sources you need are the IRS 990 database and state SOS filings.

For 990 data, the IRS 990 Nonprofit Data actor pulls Form 990 records by EIN, organization name, or state. You can query by state and revenue range, then filter on preparer fields. The actor returns structured JSON with revenue, expenses, preparer name, preparer EIN, and filing year — everything you need for first-pass screening.

For SOS filings, the actor depends on the state you're targeting:

• California: California Secretary of State
• Texas: Texas Secretary of State
• Florida: Florida Secretary of State

Each actor returns entity type, registration date, registered agent, and status. Filtering for "nonprofit" entity types incorporated in the last 24–36 months gives you the freshest organizations to target before competitors do.

What the Filtering Logic Looks Like

Here's the screening logic in plain terms:

High-priority prospects:

• 990 revenue between $200K and $3M
• No paid preparer listed, or preparer is not a licensed CPA firm
• Organization incorporated 2–5 years ago (past survival risk, stable enough to budget for services)
• Program expense ratio above 65% (financially healthy, good optics for donors)

Medium-priority:

• Preparer is a small sole practitioner (potential to displace with better service)
• Revenue growing YoY by more than 20% (approaching thresholds that require more sophisticated tax work)
• New incorporation with no 990 on file yet (reach out early, offer first-year setup help)

Skip:

• Organizations with Big 4 or regional firm preparers (high switching cost)
• Revenue under $100K (too small to afford professional fees)
• Organizations in dissolution or delinquent status per SOS records

Adding Contact Information

990 filings include officer names and titles. The executive director or CFO is usually the decision-maker for accounting services. Cross-referencing the organization name with YellowPages Scraper often surfaces a phone number or website.

From there, you have a name, an organization, their financial profile, and a contact point. That's enough to write a personalized outreach email without any manual research.

The Business Case for Automating This

A CPA firm running this process weekly can generate 15–30 qualified nonprofit prospects per state per month with essentially zero manual effort after initial setup. At a typical accounting engagement of $5K–$15K per year for a mid-size nonprofit, one new client per month from automated prospecting pays for the tooling many times over.

The firms that will dominate nonprofit accounting in the next five years are the ones building systematic prospecting pipelines now. The data is public. The question is whether you build the pipeline before your competitors do.

Practical Notes on Data Quality

A few things worth knowing before building this:

IRS 990 data is updated periodically, not in real time. The most recent returns available are typically 12–18 months old. Plan your outreach cadence around this lag.

SOS data quality varies by state. California's SOS database is comprehensive but has anti-bot protections (requires API key). Texas and Florida are more accessible. If you're multi-state, start with Texas and Florida, then add others once the workflow is validated.

Not all nonprofits file 990s. Organizations with gross receipts under $50K file a 990-N (postcard), which has minimal data. Focus on the $200K+ revenue band where full 990s are available and the data is rich enough to screen effectively.

The goal isn't to replace judgment with automation. It's to spend your judgment on the 30 organizations that actually fit your ideal client profile, rather than the 3,000 you'd have to sift through manually.