Identify Your Prospects

If you already use Claude, ChatGPT, or another AI assistant, you can connect it directly to Local News Catalyst's funder database. That gives you a permanent research connection — one you can come back to again and again to identify new funding opportunities and test ideas and get new insights using real grant data.

This page is for people who want to go further than the Funder Tool on its own allows: customizing the angle, exploring trends, or building an ongoing watchlist that fits how your station actually works.

What you're connecting to

The same dataset that powers the Funder Tool: five years of media-related grantmaking from across the country — tens of thousands of grants, each with the funder, recipient, amount, and date of support. The connection is read-only. Nothing your AI does on your end changes the underlying data, and nothing about your station, prompts, or conversations is sent back to Local News Catalyst.

Why connect your AI to a database at all

When you ask ChatGPT or Claude about funders today, it answers from general training data — which is months or years out of date and full of guesses. Connecting to a live database means the AI is reading the actual grant records when it answers you. Ask "which foundations have funded rural public radio in the last two years?" and instead of a plausible-sounding list, you get one grounded in real grants.

The technical name for this is MCP — Model Context Protocol — a standard way for AI assistants to query external data sources. You don't need to understand the protocol to use it. You set up the connection once, and from then on your AI knows it has the database available whenever a question calls for it.

A note on privacy and security

A few things worth knowing before you connect:

The connection is read-only. Your AI can query the grant data; it cannot change it, and it has no access to anything else on the LNC site.
Your conversations stay in your AI tool. What you ask Claude or ChatGPT is governed by their privacy policies, not ours. Review your AI provider's settings if you're discussing anything sensitive about your station.
Treat AI output as a draft, not a final answer. Always verify a funder's current priorities and application requirements directly on their website before reaching out.

How to set it up

Pick the tool you use. The server URL is the same for all three:

https://lnc-funder-mcp.netlify.app/mcp

Claude (Pro, Team, or Enterprise)

Go to Settings → Customize → Connectors
Click + → Add custom connector
Name it LNC Funder Database, paste the URL above, and click Add
To use it in a conversation, click + → Connectors and toggle it on

ChatGPT (Plus, Pro, or Team)

Go to Settings → Connectors (or Tools)
Select Add connector → Custom
Paste the URL above and save

Cursor, Windsurf, or other MCP-compatible editors

Add this to your MCP configuration file:

{
  "mcpServers": {
    "lnc-funder-db": {
      "url": "https://lnc-funder-mcp.netlify.app/mcp"
    }
  }
}

Start here: build your own funder list

The most direct use is replicating what the Funder Tool does, but with full control over the framing. Once you have the connection set up, try a prompt along these lines (filling in your station's specifics):

I work at [station name], a small public [radio] station in [region], serving roughly [population] across [counties or area]. We focus on [local news / arts / Indigenous coverage / rural issues — whatever fits]. Using the LNC funder database, identify foundations that have made media-related grants that look like a plausible fit for us. Group them into: (1) funders who have supported similar stations, (2) regional or place-based funders likely to care about our coverage area, and (3) topic-aligned national funders. For each, give me funder name, amount range of their typical media grants, and a brief reason they're worth considering.

A few things this prompt does that a generic AI question won't:

It anchors the AI in your actual context, not assumptions
It explicitly asks the AI to use the database (you can usually tell whether it did by whether the funders feel specific and verifiable)
It asks for organized output rather than a flat list — easier to act on

Going further with the same connection

Once the connection is set up, it stays. Here are some other ways stations have used live database access:

Spot funding trends

Ask the AI to look at year-over-year shifts in who's funding public media, or what kinds of work are getting funded more (or less). For example:

Which foundations have increased their media funding most over the last three years, and where is that growth concentrated geographically or topically?

Build a watchlist

Identify ten or fifteen funders worth tracking, and revisit the conversation every few months to see what they've funded since. The connection means the data is always current.

Here's my watchlist of fifteen funders. What have they each funded in the last two years that involves local journalism, rural communities, or public broadcasting? Flag anything that looks newly active.

Learn from peers

Pull the funding history of a few stations you consider peers and look for patterns: who funds them, at what amounts, and how often. This is often more revealing than a generic search.

Pull the grants received by Station A, Station B, and Station C over the last five years. What funders show up across more than one of them, and what does the typical grant size look like?

Prepare for a funder meeting

Before a call with a program officer, pull what their foundation has funded recently to ground your conversation in their actual priorities rather than their stated ones.

Show me the last twenty media grants made by the [Foundation Name]. What patterns do you see in size, geography, and recipient type? What questions should I be prepared for if I'm pitching them?

When you're ready to write

Identifying funders is step one. When you're ready to approach them, we have tools to help you build a letter of inquiry or develop a full application — both grounded in the same kind of station-specific data.

Go to the Secure New Grants page to develop the materials you need.