As you answer the AI's questions, you can upload documents or paste in information to help it understand your work more deeply. You don’t need everything listed below; even a few key details will make a real difference to the quality of the output.

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

• Your mission and the people you support, in plain language
• Your current fundraising income streams (if any) - e.g. grants, events, individual giving
• Your capacity: rough staff and volunteer time available for fundraising
• Any hard constraints: no in-person events, restricted to a specific geography, etc.
• What kind of supporters you have: local community, corporate partners, individual donors, etc.

Additional context to add depth and detail

• Previous fundraising activity: what you’ve tried and how it went
• Your organisation’s tone and values: what feels right for you
• Any funding gaps you’re trying to fill: unrestricted income, capital, etc.
• Seasonal considerations: events, awareness days, your community’s calendar
• Partnerships or networks you already have: local businesses, schools, faith groups
• Your latest annual or impact report

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.

You’ll receive a list of ideas with varying approaches and effort levels. The AI may ask clarifying questions first, for example about your community or whether certain activities are realistic for you. Answer honestly: it leads to better suggestions.

Not every idea will be the right fit, and that’s fine. Think of this as a starting menu, not a to-do list. Your job at this stage is to evaluate, not execute.

Questions to ask yourself as you review:

•  Do any of these ideas feel genuinely exciting or within reach?
• Are there ideas here that we could test on a small scale first?
• Do the ideas reflect what our supporters actually care about?
• Are the effort estimates realistic for our team right now?
• Are there any ideas that could feel uncomfortable or inappropriate given who we work with?
• Do the ideas represent the breadth of our supporter base, not just the most obvious audience?
• Does anything feel too generic? If so, you can ask for it to be tailored more specifically.

 If something doesn’t feel right, that’s fine: you can refine it in the next step.

Once you have a starting list, you can go deeper on the ideas that look most promising. Think of this as a conversation - keep going until you have something genuinely useful.

Ask for revisions - example prompts as below:

Tone:

• “These ideas feel quite standard. Can you suggest something more distinctive that would stand out?”

Length:

•  “We’re a very small team. Can you filter these down to the three most achievable options?”

Structure:

•  “Create a simple comparison table of these ideas across effort, cost, and likely income potential.”
• “Take the top three ideas and draft a one-page summary of each that I could share with our trustee board.”

Representation:

• “Which of these ideas could be led or co-created by the people we support?”
•  “Can you suggest ideas that don’t rely on ‘need’ or hardship narratives to appeal to donors?”

If this list doesn’t feel right, you can always start a fresh conversation with slightly different instructions; sometimes that’s a faster route to ideas that fit better.

Add what only you know

AI can’t know your local community, your relationships with supporters, or the history that makes certain ideas resonate for your organisation. Once you have a shortlist of ideas you like, make sure to layer in:

•  Specific local connections, venues or networks you could draw on
• Your organisation’s personality and what feels true to your culture
• Existing supporter enthusiasm or momentum around particular themes
•  Budget realities and any upcoming constraints on your team’s time

Use this as a team exercise and conversation starter

The ideas AI generates can be a useful prompt for a team or trustee discussion. You don’t have to take the AI’s output at face value. Sharing it with colleagues and asking “what resonates and what doesn’t?” is often where some of the real thinking happens.

Test before you commit

Before investing significant time or money in any idea, ask AI to help you design a low-cost test or pilot. Many fundraising ideas can be trialled at a small scale - with a single event, a small donor cohort, or one local business -  before you go further.

Be honest about ethics

Some fundraising approaches can, unintentionally, sometimes frame your beneficiaries’  lives in ways they wouldn’t recognise. The REACH harm check helps flag this, but it’s also worth asking your team and, ideally, the people you work with whether the approach feels right.

Always review before acting

AI-generated fundraising ideas can be a great springboard, but they need human judgement before you act on them. Check income projections against your own knowledge, run ideas past someone with fundraising experience, and never commit to a plan based solely on what AI suggests.

Be transparent about AI use

If you use AI to generate ideas that shape your fundraising strategy, consider noting this in internal records or reporting. Being open about how AI tools support your work is part of responsible practice.