You run the programs, write the grants, thank donors, fix the printer, and close the books, sometimes all in the same afternoon. Most advice about AI for small nonprofits assumes you have a tech team and a budget line. You have neither, and you do not need either to start. Here is exactly where to start if you are curious about AI, a five-step plan that costs nothing, and what to skip entirely.
In this post: the three places I suggest that small nonprofits start, a five-step plan, what to skip, what it costs, how to stay safe without an IT department, and why it is worth your time to see if and how AI can help you.
The three places I suggest small nonprofits start with
I encourage you to start where the stakes are low and the repetition is high. That rules out anything touching client data and points straight at the writing your week is already full of.
Donor communications come first. Thank you letters, drafts of appeals, newsletter content, and lapsed donor notes are high volume, low risk, and involve no client data. I wrote a full post on why donor communications are where most nonprofits should start with AI, and everything in that post applies double to a small shop.
Board and funder reporting comes second. Give an AI assistant copies of a few recent reports as writing samples, your program numbers and bullet points you want included, and it will produce a readable first draft of the narrative you rewrite every quarter. You edit, you approve, you send. The judgment stays with you, and the blank page problem goes away while you save time.
Meeting notes are third. Record your staff or committee meeting or use one of the countless note-taking apps on the market, get a transcript, and have AI summarize decisions and action items. The follow-up email that used to take forever takes ten minutes.
Notice what all three have in common: a human reads every word before it goes anywhere, and no client information ever enters the tool.
Your five-step plan with AI: a walk-through
Here is what a realistic five-step plan looks like. No new software, no budget request.
Step one, pick one task: this month’s donor thank you letters. Open a general AI assistant and write a short instruction in plain English: who your organization is, who the donor is, what they gave, and the warm, plain tone you want. Paste in your old letter as an example of your voice.
Step two, read the drafts of the donor letters critically. They will be 80 percent right and 20 percent generic. Fix the 20 percent. Notice what you fixed, because that becomes part of the instructions you provide to the AI assistant next time.
Step three, do it again with the next group of letters. Your instructions improve, the drafts improve, and the whole batch takes 40 minutes instead of three hours.
Step four, try to generate a program report for your board: attach two or three previous reports as writing samples, paste in your program stats and bullet points that you want highlighted, and ask for a one-page board update in your voice.
Step five, write down what you learned as three rules for yourself: what you will use AI for, what never goes into it, and the rule that every word gets read before it goes out. Congratulations, you have just started your AI policy.
What to skip
Wait on anything client-facing until you have real experience that will guide you and your team. A chatbot answering client questions can go wrong in ways a draft thank you letter cannot. Done properly, though, client-facing AI is some of the best work happening in the sector: our information and referral chatbot case study shows what it looks like when it is built carefully, with staff in control. Wait on it; do not write it off.
Skip AI in the hiring process entirely. Screening tools carry bias risks that a small organization has no way to audit. While you may save time by using AI to screen resumes, there is a very real risk that you will miss out on good candidates. I wrote a full post on why nonprofits should keep AI out of the hiring decision if you want the full case.
Skip anything that requires moving your data into a new platform or shelling out significant money until you truly feel that the small steps you are taking with AI are benefiting you and your organization. The tools that work inside of the applications you already have, like Google Workspace or Microsoft 365, are cheaper and safer than a new system you have to migrate into. And skip any vendor who opens with a discount that expires Friday. A tool worth buying will still be worth buying next month.
What it costs
Less than you fear. The major AI assistants have free tiers that are enough for a small organization just starting out. Paid tiers generally run about the cost of a modest software subscription per person, and some vendors offer nonprofit discounts. Check current pricing directly, because it changes often, and never let a tool’s price be the reason you skip the data questions in the next section.
The real cost is your time and attention: a few hours to learn what good instructions look like, and the discipline to review everything. Budget your time, not just your dollars. The investment of time and attention can be significant up front. However, you begin to realize time saved quickly.
Staying safe without an IT department
Be clear about what never should be shared with AI. Client names, stories that have names and highly personal information attached to them, and anything you would not put on a postcard should stay out of AI tools. This habit can prevent the worst outcomes from being realized.
Use a work account, not a personal one. Organizational accounts on the major platforms come with settings that keep your data out of model training. Personal free accounts often do not.
Write the rules down. Even a one-page policy beats none, and your week of experiments already gave you the first draft. When you are ready to make it real, my post on what your nonprofit AI use policy should actually cover walks through every section.
Why it is worth your time
AI cannot know and understand your community, build relationships, or make a judgment call. It has no idea why Mrs. Alvarez gives every December or what your board chair meant by that email. The work that requires judgment and human interaction stays yours.
Every hour AI gives back to you flows into the mission and allows you to do more. This is the order of priorities I bring to every AI decision: mission first, staff second, the people you serve third, and the tool last. The administrative load is a major driver of burnout in our sector. This is especially true for smaller organizations.
Common Questions
Can a small nonprofit afford AI?
Yes. The free tiers of major AI assistants cover most of what a small organization needs in its first year: drafting, summarizing, and formatting. The real investment is a few hours of your attention to learn what good instructions look like.
What is the best way for a small nonprofit to start with AI?
Start with donor thank you letters or another high volume, low risk writing task that involves no client data. Get good at one task before adding a second. One useful habit beats five abandoned experiments.
Do small nonprofits need an AI policy?
Yes, and a one-page policy is fine to start. Write down which tools you use, what information never goes into them, and the rule that a person reviews every output. Grow the document as your use grows.
Will AI replace staff at small nonprofits?
No. Small nonprofits have too few people for too much work, and AI absorbs busywork, not relationships or judgment. The realistic outcome is that your existing team gets hours back for the mission, not that anyone becomes replaceable.
If you are the ED wearing every hat, you do not need another tech lecture. You need an hour of honest talk about where AI fits your organization. I am glad to talk it through with you, no jargon, just your questions and straight answers. Book a short call with me at calendly.com/larry-nonprofitnext/30min.
Larry is the founder and Principal Innovation Strategist at NonprofitNext. Learn more at nonprofitnext.ai.