On starting before you feel ready, and why people & impact come before tools.
If you work in a mission-driven organisation, you have probably felt the pressure by now. AI is in every newsletter, every funder conversation, every conference agenda. The quiet worry underneath it is that everyone else has worked this out and you are behind.
I want to offer a different starting point, because I think the pressure is pointed in the wrong direction.
The goal was never to use AI. The goal is impact. For most nonprofits and education organisations, that means doing meaningful work with genuinely limited resources, and the most limited resource of all is usually time. AI is interesting only insofar as it helps with that. It is a means. It is not the end or point. When “adopt AI” becomes the objective itself, you end up with tools in search of a problem, and that is how good intentions turn into shelfware. For example, the number of tokens used as a measure of AI’s impact is a poor one when compare to the actual difference it makes in time and money saved.
So the first question is not “which AI tool should we use?” It is “where is our time actually going, and where does it hurt?”
That sounds obvious, but it is the step most often skipped. I have seen teams reach for a tool before they could clearly describe the problem they wanted it to solve. The result is predictable. The tool gets configured, a few people are trained, and within a month it is quietly abandoned, because it was never attached to a real need in the first place. The technology was fine. The sequencing was wrong.
Understanding the need takes some patience. It means talking to the people doing the work and asking where the friction is. Where does information go missing? Which tasks get done three times by three people because nobody could see the others doing them? What takes a whole afternoon that everyone has simply accepted as normal? You are not looking for AI use cases. You are looking for honest answers about how work actually happens. The AI part comes later, and it comes more easily, once you know what you are aiming at. I have written separately about how to sequence that work in phases, in a piece for the Atlassian Community, if a more structured approach would help.
Here is the part I most want to say to anyone who feels behind. Start anyway. You do not need to feel certain. You do not need a strategy document or a budget line or a confident plan. You need one real problem and the willingness to try something small against it. A basic plan written on the proverbial napkin is fine. It is important for having a sense of direction and orientation. The plan should not be followed at the cost of flexibility, though.
Understanding deepens by doing, not by waiting. The teams that make progress are not the ones who felt ready. They are the ones who picked a genuine pain point, tried something modest, and paid attention to what happened.
When you work in this order, needs first and tools second, AI stops being intimidating. It becomes what it should have been all along: one useful way to give limited resources more reach. A way to spend less time hunting for information and more time on the work only people can do. That is not a small thing. For an organisation stretched thin, magnifying impact without magnifying cost is close to everything.
So if you are feeling the pressure, set the pressure down. You are not behind. You are just being asked the wrong question. Start with a real human need, start before you feel ready, and let the tools follow. That sequence has never let me down.