Why nonprofits can’t afford to wait on AI

Hi, I'm Johnny.

I've spent time in both worlds.

Startups racing to jump on the AI train and nonprofits doing essential work.

But what's striking to me is not the difference in talent or how much people care, but it's the difference in how each side approach AI.

In the startup world, someone's asking every day, what can AI do for us?

How can we streamline this?

In a lot of nonprofits I've worked with or supported, the question is still, is AI even relevant to our work?

And my answer is, yes, it absolutely is.

And the window to get ahead is now.

So in this video, I will, talk talk about how you can run a simple 30 day AI pilot at your nonprofit to create capacity with the team you already have.

Not a big strategy, not a huge project, just a focused month to get real practical wins.

Most nonprofits I talk to or work with are in the same spot.

You're stretched thin.

You're watching every dollar.

You have a list of things you wish you could do if if you had just more time or more people.

You've You've probably also heard a lot of noise about AI, but you don't have hours to sit through webinars or figure out where to start.

So the goal of this 30 day pilot is not to automate everything or replace people.

It's to reclaim a few hours a week from tedious, repetitive work and prove to yourself that AI can be a real capacity multiplier for your mission.

So week one is about choosing tools and your scope.

By the end of week one you should be able to answer the two questions which AI tools are you going to use and for which specific tasks?

For most organizations I recommend starting with one general purpose AI agent like Anthropic's, Claude or tools from OpenAI.

You might also add one AI enabled tool that you already use like your expense platform or your CRM.

if it has AI built in, then pick on just two workflows to focus on.

Look for work that are repetitive, text heavy and currently done by humans.

Some common examples of that is drafting up grant reports or donor updates, Summarizing program notes and impact for donors drafting email campaigns, newsletters or social posts.

Don't overthink it.

By by the end of week one you'll want a simple sentence like this: "We're going to use Claude and our existing tools to help with grant reports and donor emails." Week two is where you give people time and space to experiment.

This is not another lunch and learn.

The idea is to block off a half day or a full day where a small group can step away from daily work and really try things.

Ask each person to show up with a real task that they need to do this week and the data or documents they would normally use to do it during this time.

Have people use AI to draft first versions of their reports or emails, or or ask AI to clean or organize notes and spreadsheets or your Salesforce.

Turn long documents into summaries or bullet points for funders, boards or staff and set expectations clearly, we're not looking for perfection.

We are looking for this got me to 60 to 80% of the way there.

In much less time.

Have everyone jot down very simply what worked, what felt confusing or tricky and where it obviously saved time.

Now Week 3 It's about sharing and tailoring by team.

Now take what you learned in experiments and turn it into real team specific practices.

Host a 45 45 to 60 minute share out.

Ask each person to show one workflow where AI actually helped.

Then group examples by function development and fundraising programs and service delivery, operations and finance, communications and marketing.

For each team.

Answer two questions together.

First, What are one or two tasks we are officially comfortable doing with AI now?

Second, what are our guardrails?

For example, No external facing content goes out without a staff member reviewing it.

We don't put sensitive data data into public tools.

Capture all of this in a short shared document.

Just a clear living note that people can come back to.

Now Week 4 is about ownership and momentum.

If you stop here, the pilot just becomes a nice one off project and that slowly fades away.

To avoid that, you need to do two things.

First, Name an AI trailblazer.

This doesn't have to be a new hire.

It's someone on your existing team who is curious about AI and has the trust of their colleagues and can spend a little bit of time going deeper.

Their job is not to be the AI department.

Their job is to keep an eye on a small number of key tools and help colleagues redefine prompts and workflows.

Collect wins, risks and lessons learned as people keep experimenting.

Second, create a home for the ongoing conversation.

That could be a Slack channel or Teams channel for AI experiments and ideas or a simple shared doc where people can drop use cases, prompts they use and questions.

The goal is to make AI part of how you work, not a one time workshop that everyone forgets.

If you run through this 30 day pilot, you should finish with three things one a small set of tools your team is actually comfortable using two A handful of real workflows where AI reliably saves you time without compromising quality or trust and three a named person and shared space to keep learning and improving over time.

you don't need a perfect AI strategy to start.

You just need one month, a few real tasks, and the willingness to experiment.

If you're not sure where to begin, pick one recurring task that your team dreads and give this 30 day pilot a try.

your mission is too important to leave AI on the table.