As you answer the AI’s questions, you can upload documents or paste in information to help it understand your work more deeply. You can paste information directly into the chat. You do not need tidy documents. A bullet-point list of your key programmes, a paragraph about your beneficiaries, or a copy of your mission statement all help the AI understand your work.

Important context (add as much of this as you can)

• Your mission and the people you support, including any demographic, geographic, or community focus
• What the funding is for: a specific project, core costs, a new service, capital spend
• Your organisation's size: approximate income, staff, and years operating
• Your location: where you work and whether your reach is local, regional, or national
• Any eligibility constraints you already know about, e.g. you must be a registered charity, CIC, or community group

Additional context to add depth and detail

• Previous funders: who has supported you before and whether those relationships are ongoing
• Funding history: whether you have applied to particular funders and not heard back
• Programmes or outcomes: what your work achieves and how you measure it
• Partnerships: other organisations you work alongside that may open up joint funding routes

Before you upload files: check for sensitive data

Review your documents for personal or sensitive information about beneficiaries. Anonymise names and identifying details before uploading. Even though Worthwhile Chat is privacy-first, good data practice starts with you. If you're using other AI tools, check their data settings too — many train on your data by default. How much you share depends on your existing privacy and data policies, and how much control you have over the tools you use.

If you’re unsure, ask yourself: “If this document were seen by someone outside our organisation, could anyone be identified or harmed?” If yes, anonymise first. 

The AI will produce a suggested shortlist of funder types and funding streams, grouped by category. It may also ask you clarifying questions, for example, whether your work covers a particular theme that could open up additional routes, or whether you have a trading income that some funders exclude.

Before you act on the output, work through these questions:

• Does the list reflect what you actually know about the funding landscape in your area?
• Are there obvious funders you know that haven't appeared? If so, ask why - the answer may reveal something about your eligibility or framing.
• Are the funder descriptions accurate? Cross-check anything unfamiliar before investing time in an application.
• Does the language used to describe your beneficiaries feel right? If not, correct it before sharing the output with anyone.
• Have any large or well-known funders been excluded, and do you understand why?

Grant research is rarely a one-prompt job. Think of this as an iterative conversation: you can keep refining until you have something genuinely useful to act on.

Ask for revisions

If the first draft isn’t quite right, try prompts like:

• Tone: "Make this sound more like advice from someone who knows our sector, not a directory listing."
• Length: "This is too long to act on. Can you narrow it down to the five strongest options?"
• Structure: "Can you group these by deadline type instead, rolling versus fixed rounds?"
• Representation: "Check how you’ve described our beneficiaries here, and flag anything that sounds outdated or stereotypical."

If the output isn't landing quite right, start fresh with a slightly different framing. Sometimes describing your work from a different angle, by outcome rather than activity, for example, surfaces different funding routes.

Add what only you know

The AI works from general knowledge of the funding landscape. If your tool doesn't search the web, treat this as a well-informed starting list. If it does, treat anything it's found as a head start, not a confirmed fact. Once you have a useful shortlist, you will need to:

• Check funder websites directly to confirm eligibility, deadlines, and current priorities
• Use specialist databases such as Funding Central, GrantFinder or Charity Excellence to verify and expand the list
• Apply your own knowledge of your local and sector-specific funding landscape, particularly for community trusts, local authority grants, and regional funders

Always verify before you apply

Even AI tools with web search can miss recent changes or misread a page. Priorities change, programmes close, and eligibility criteria shift faster than any single check can guarantee. 

Always check the funder's own website before investing time in an application. Treat the AI output as a starting point for research, not a confirmed list. 

Be specific about your organisation

Vague prompts produce vague results. The more specific you are about your income, location, beneficiaries, and what the funding is for, the more useful the shortlist will be. If you are exploring funding for more than one project, run separate conversations for each. Mixing multiple needs in one prompt tends to blur the output.

Check the language used about your beneficiaries

AI tools can default to language that is clinical, outdated, or disempowering when describing the people charities support. Review any funder-facing language in the output carefully. If something doesn't feel right, ask the AI to reframe it, or simply replace it yourself.

Be transparent about AI use

If you share a funding landscape summary or pipeline document with colleagues, trustees, or partners, consider noting that it was produced with AI assistance and has been reviewed by a member of staff. Being open about AI use builds trust and helps others understand the limitations of the output.

A note on human review

Never share AI-generated funding research externally without reviewing it first. Funder descriptions should always be verified against primary sources. A colleague familiar with your funding landscape is an invaluable sense-check before you commit to a research plan.