For every 100 open, source-tagged grant records in the GrantLens database, about 85 come from federal sources and about 15 come from state sources. That is the practical starting point for grant writers in Miami-Dade, Broward, Palm Beach, Alameda, Contra Costa, Marin, and San Francisco counties: a local grant search often needs a federal research strategy behind it.
That does not mean every nonprofit should pursue every federal opportunity. It means that limiting a pipeline to local-looking opportunities can leave out most of the currently open source-tagged inventory. The better question is not whether a funder is federal or local. It is whether the program's eligible applicant, geography, activity, partnership requirements, award size, and reporting burden match the organization in front of you.
Federal sources dominate the open-grant inventory
The database currently contains 1,294 open federal-source records and 221 state-source records. Federal records therefore make up 85.4% of the 1,515 open records with a federal or state source label. Yet the source mix is not the same thing as the money mix: state-source records carry a larger total dollar amount in this dataset.
View data
| Open grant records | |
|---|---|
| Federal | 1294 |
| State | 221 |
| Source | Open grants | Total amount | Average amount per open grant |
|---|---|---|---|
| Federal | 1,294 | $1,268,399,352 | about $980,216 |
| State | 221 | $3,160,344,695 | about $14,300,202 |
The average state-source amount is roughly 14.6 times the average federal-source amount. An average is not a recommended request size, and it is especially vulnerable to a small number of very large awards. Still, it gives development teams an important planning signal: federal searching may produce more individual possibilities, while a smaller state-source set may include unusually large, complex opportunities.
For a South Florida or Bay Area nonprofit, that distinction should shape the way staff allocate time. A lean team may want a broad federal search for programs that fit its core services, then reserve substantial internal capacity for the occasional larger opportunity that requires public-sector partnerships, detailed compliance systems, or a longer pre-application process. Use filters to search by source and geography, but do not treat county relevance as proof of eligibility. A grant can appear useful to a local organization while still being limited to governments, institutions, or another applicant type.
The historical federal award mix is concentrated
The historical award data also shows why a source-only search is too broad. Among the leading federal funders in records with reported amounts, the largest funder accounts for 44.2% of total awards. The top two account for 66.1%, and the top three account for 74.7%. The top 10 together represent 96.2%.
| Rank among leading federal funders | Share of total historical awards |
|---|---|
| 1 | 44.2% |
| 2 | 21.9% |
| 3 | 8.6% |
| 4 | 6.6% |
| 5 | 3.2% |
| 6 | 3.1% |
| 7 | 3.0% |
| 8 | 3.0% |
| 9 | 1.6% |
| 10 | 1.0% |
Concentration changes the research task. If a nonprofit sees a large number of federal results, the right response is not to skim them in order of deadline. It is to identify which funding systems repeatedly support the kind of work the organization actually does, then learn their patterns: typical recipient profile, award scale, repeat grantees, partner expectations, and whether the work is direct service, infrastructure, research, planning, or pass-through implementation.
Historical awards show a directional shift as well. The number of awards fell from 2,257 in 2024 to 1,563 in 2025, a decline of about 30.7%. Total awards declined about 25.8%, from $3.769 billion to $2.797 billion, while the average award rose about 7.2%, from $1.670 million to $1.789 million. That combination can mean fewer awards do not automatically mean less money; it may mean a more selective landscape with larger average awards.
Before assigning a proposal to a busy program team, review funder records and award patterns. Looking at prior recipients is not about copying a successful organization. It is about recognizing the operating model that appears competitive: a standalone nonprofit, a public agency partner, a multi-organization coalition, or an entity with a specialized delivery role.
Category totals reveal a mismatch between averages and typical awards
Open-grant categories are not equally sized, and their averages can conceal very different award distributions. Community development has the highest reported average amount at $51.033 million, but its median is $1.5 million. Social services has 655 open grants, the most of any category, with a $28.385 million average and a $1 million median. Those gaps are a warning against using the average award as a budget target.
| Category | Open grants | Average amount | Median amount |
|---|---|---|---|
| Social services | 655 | $28,385,214 | $1,000,000 |
| Health | 580 | $2,371,707 | $500,000 |
| Technology | 250 | $10,811,616 | $750,000 |
| Education | 236 | $14,412,117 | $112,500 |
| Environment | 145 | $6,931,253 | $750,000 |
| Community development | 124 | $51,033,379 | $1,500,000 |
| Arts and culture | 123 | $479,944 | $25,000 |
| Public safety | 66 | $2,701,286 | $1,460,000 |
| Food security | 52 | $2,485,854 | $750,000 |
| Transportation | 23 | $16,000,000 | $5,000,000 |
For executive directors, the median is often the more useful early planning figure because it is less influenced by unusually large awards. For grant writers, the gap between average and median is a prompt to read the funding notice closely. A large maximum award may be intended for an applicant type, scope, or consortium that does not resemble your organization.
This is especially important for organizations that deliver community development, arts and culture, or social services in the listed South Florida and Bay Area counties. The search should begin with mission fit, then narrow by applicant type and implementation capacity. A small arts organization should not assume a community-development result is too large to consider, but neither should it build a budget around the category average without seeing the actual program rules.
A few currently open listings illustrate why labels alone are insufficient. Federal Disaster Assistance is described as federal financial assistance for recovery following a declared federal disaster, with eligible applicants including cities, counties, state agencies, and certain private nonprofits. That is a potentially meaningful distinction from a general operating grant: the triggering event, applicant status, and recovery purpose are central to fit.
Likewise, the FFY 2024 State & Local Cybersecurity Grant – Local & Tribal (SL) RFP is aimed at local and tribal government entities. A nonprofit that sees a technology-related search result should therefore verify the applicant definition before investing time in a narrative. The fastest way to avoid a preventable no-go decision is to check eligibility and readiness gaps against the requirements that matter: entity type, tax status, geography, required partnerships, registrations, and capacity to manage the award.
Local searches need a source strategy, not a source preference
A nonprofit in Miami-Dade or San Francisco may begin with a county-specific need: youth programming, family supports, public-space activation, community development, technology capacity, or arts access. That local framing is appropriate. The mistake is assuming that a local need requires a local funder, or that a federal source necessarily funds only national work.
Build each search around the work rather than the source label:
- Define the funded unit of work. Is the request for direct services, staff training, planning, equipment, capital work, data systems, public outreach, or a partnership?
- State the applicant model. Decide whether your organization can apply directly, must work through an eligible public entity, or should serve as an implementation partner.
- Set a realistic award band. Use category medians as a rough orientation, then let the actual notice determine the request amount and period of performance.
- Separate discovery from qualification. Save promising results even when a requirement is unclear, but do not advance them into proposal production until eligibility is resolved.
- Assign an owner and a decision date. A grant is not a pipeline asset if no one is responsible for reading the notice, contacting partners, or deciding whether to proceed.
This approach also helps teams recognize when a state or non-federal listing belongs in the same pipeline. For example, the Local Parks Grant Program is described as supporting art and cultural projects that strengthen community ties to local history and natural spaces. The General Program Support - Local Arts Agency listing has a stated maximum of up to $58,500 and is limited to local arts agencies. Both examples underscore the same discipline: category fit is only the opening screen; applicant eligibility and program purpose determine whether a result is actionable.
When a notice is long or written for multiple applicant types, turn it into a working document rather than relying on memory. Teams can turn a NOFO into a requirements checklist and use the resulting list to assign evidence, attachments, approvals, and open questions before a deadline becomes urgent.
What this data cannot tell you
The figures above are a useful view of the GrantLens database, but they are not a census of every available grant. All open-grant figures reflect the database at generation time rather than the full universe of opportunities. An opportunity may be newly posted, changed, closed, or absent from this snapshot.
The dataset also has no open-grant county data. It cannot tell you how many open grants are available specifically in Miami-Dade, Broward, Palm Beach, Alameda, Contra Costa, Marin, or San Francisco counties. Geographic filtering remains essential because an organization’s service area can affect eligibility, competitiveness, partnership options, and the practical value of a grant.
Award-history coverage varies by source and year. For that reason, the movement from 2024 to 2025 should be treated as directional, not as a definitive statement about the entire funding market. The historical totals also do not reveal the quality of applications, whether awards were renewed, the timing of payments, indirect-cost treatment, or the staff time required to administer a grant.
Finally, averages and medians describe records, not organizational fit. They cannot tell a development director whether a $500,000 opportunity is manageable for a particular nonprofit, whether a board will accept the compliance obligations, or whether a proposed program has the evidence and partnerships needed to compete.
Build a federal search pipeline around fit
The federal share of the open inventory makes federal research hard to ignore: about 85 out of every 100 source-tagged open records are federal. But the concentration of historical awards, the wide gaps between category averages and medians, and the absence of county-level open-grant counts all point to the same operating rule: filter aggressively before writing.
For nonprofits serving South Florida or the Bay Area, start with the service area and mission, screen applicant eligibility, examine comparable award patterns, and only then commit proposal time. Build a filtered grant search that reflects your organization’s work, capacity, and real geographic reach.