Thinking Strategically About AI
A working guide for small nonprofit executive directors who cannot afford another hat.
Your answers save automatically in this browser - nothing is sent anywhere, and no email is required. Close the tab, come back next week, and your work will still be here.
AI strategy is your job. Good news: you already know how to do it.
You set direction for your board. You decide which programs live and which get sunset. You choose where every scarce dollar goes. That is strategic thinking, and you do it every week.
Deciding how your organization uses AI is the same kind of decision. It is not a tech task to hand to the youngest staffer because they know Canva. It is a leadership call about where your people's time goes - and time is the scarcest thing you have.
Strategy asks why. Operations asks how.
Strategic thinking about AI means deciding which problems are worth pointing AI at and why. Operational work is figuring out which buttons to press. Most organizations skip straight to the buttons, try a chatbot for twenty minutes, shrug, and walk away thinking AI is just Google with better manners.
Get the why right first, and the how gets dramatically easier. That is what this guide helps you do.
A reframe worth keeping: adopting AI is not admitting you can't keep up. It is the opposite. "I built systems that don't depend on any one person" is one of the strongest sentences an executive director can say to a board. This guide helps you earn it.
How to use this guide
Seven short sections, each with something to read and something to do. Work through it in one sitting (about 40 minutes) or one tab at a time over a week. Your answers build toward a one-page AI Strategy Snapshot you can print and bring to your board.
Your constraints are real. So are the openings they create.
Small nonprofits face pressures that big organizations write white papers about. Here is what changes when you look at each one through an AI lens.
Quick readiness check
Check every statement that is true for your organization today. Be honest - this is for you, not your board.
You do not need a 40-page strategic plan. You need the right-sized approach.
Strategic planning models fill textbooks. For a small nonprofit thinking about AI, they boil down to four. Pick the one that matches your situation right now.
Whichever you choose, one classic tool earns its place: the SWOT analysis. It takes 15 minutes and turns "we should probably do something with AI" into an actual picture of your situation.
Your AI SWOT
Answer in plain sentences. Rough is fine - you can refine later.
Your mission is the filter. Run every AI decision through it.
Mission drift rarely announces itself. It creeps in one shiny tool at a time. Three questions keep you honest before adopting any AI use:
The ethical questions deserve real answers
In nonprofit circles, concerns about AI's energy use, water consumption, bias, and privacy come up constantly. They should. These concerns are real, and anyone who waves them away is not being straight with you.
They are also manageable. You can choose lighter-weight tools and batch your use. You can treat every AI draft as a draft from a well-read stranger who doesn't know your community - reviewed by a human who does. You can keep sensitive data out of tools that aren't approved to hold it. Taking the concerns seriously is not a reason to opt out of the conversation. It is the qualification for leading it.
The rule that covers most sins: AI drafts. Humans review. Humans publish. Every time, without exception.
Governance starter kit
Check each commitment you are ready to make. These five lines are a working AI policy - most small nonprofits need little more to start safely.
Don't start with your biggest problem. Start with your loudest groan.
The biggest problem is usually the worst first project - too tangled, too many stakeholders, too easy to stall. The right first pilot is a task that is repetitive, time-hungry, and low-risk. The kind that makes you groan when it lands on your calendar.
What a good first pilot looks like
Small: 3 to 5 hours of total effort. Measurable: you know the hours it takes now, so you can count the hours it takes after. Real: run on your actual work, not a demo. Documented: written down step by step, so it keeps working after the person who built it moves on.
Here is the pattern, using board reports as the example. Today: four hours of gathering numbers, hunting through email for that one figure, and formatting. With AI in the loop: you feed it your notes and data, it assembles a first draft in minutes, and you spend 45 minutes editing and fact-checking. You saved roughly three hours, this month and every month after. Three hours in a month may not seem like a lot, but it sure feels like a lot when that report is due.
Pick your pilot
Skip the dashboard. Track three numbers and one feeling.
Impact measurement can swallow whole staff retreats. For your first AI effort, keep it to this:
One simple rule: if the pilot does not save more time than it costs, stop. That is not failure - that is data, and it cost you almost nothing to learn. Pick a different task and try again.
Time-back calculator
Don't go it alone
AI changes fast. You do not need to keep up with all of it - you need one trusted newsletter, a couple of peer EDs to swap notes with, and a quarterly hour to review what's working. The peer piece matters most. The best AI decisions I've seen nonprofit leaders make started with a conversation, not a search bar.
Your AI Strategy Snapshot
Here’s everything you entered, gathered onto one page. Print it and bring it to your board, or stick it on the wall where the Post-it notes go to die, but don’t let it die because it’s some great stuff!
Click "Build / refresh my snapshot" and your answers from the earlier tabs will appear here.
Want a guide for the next step?
If you worked through this and thought "I could use someone to walk beside me on this journey" - that is exactly what I do. I'm MJ, and I help small nonprofits use AI in ways that serve people, purpose, and what matters most. No jargon and no pressure.
Let's talk