Using AI for better nonprofit fundraising with Josh Hirsch
5/20/2025
“It’s about knowing that you’re not going to break anything, right? So like, like I said, don’t be scared. Go into it understanding that It will not always be right the first time, and it has the potential for bias. Like, inherently, AI is biased based upon the data collected from the LLM, based upon our own initial bias.
So, identifying ways to create those feedback loops within whatever tool you’re using, Chat GPT, Gemini as such, and learning from it just as much as you are just generating straight output. So I love prompting it and asking it, you know, things like, Do you understand or have any question? By me prompting it with that, I’m able to get a better understanding if it needs clarity, if it needs more information.”
In episode 123 of Nonprofit Mission: Impact, Carol Hamilton and Josh Hirsch discuss using AI in fundraising and donor engagement.
They explore:
The evolution of AI, and its potential to enhance how organizations connect with donors.
How to train AI tools effectively, leveraging data responsibly, and using AI to enhance—rather than replace—human-driven strategy.
Why nonprofit leaders who embrace AI thoughtfully do not need to fear the technology
EPISODE HIGHLIGHTS:
The Evolution and Role of AI in Nonprofits
[00:010:00] - AI tools, such as ChatGPT, have rapidly become mainstream, presenting both opportunities and challenges for nonprofit leaders.
[00:11:07] - Understanding AI beyond ChatGPT: The importance of knowing different tools and their specific applications.
How Nonprofits Can Use AI Effectively
[00:12:31] - AI can personalize donor communications beyond basic mail merges, creating bespoke engagement strategies based on donor history and interests.
[00:14:10] - Predictive AI helps nonprofits identify donor retention risks and re-engagement opportunities before they lapse.
[00:15:40] - AI allows organizations to create customized 12-month communication plans for different donor segments, enhancing stewardship.
Training AI to Match Your Brand’s Voice
[00:16:51] - Understanding zero-shot, one-shot, and few-shot learning in AI and how they impact the quality of content generated.
[00:18:19] - How nonprofits can train AI models using their past communications, mission statements, and brand guidelines to improve accuracy.
[00:19:42] - Practical example: Using AI to generate content outlines that can then be refined into original, authentic messaging.
Best Practices for AI Implementation
[00:20:31] - AI should be seen as an enhancement, not a replacement, for nonprofit staff. Organizations that leverage AI effectively will outperform those that don’t.
[00:22:16] - Strong prompting techniques are essential for getting useful AI-generated content—specificity and clarity in prompts lead to better results.
[00:24:52] - AI is a tool that should be used with a feedback loop—nonprofits must refine and personalize AI outputs rather than using them verbatim.
Ethical Considerations and Policy Development
[00:25:25] - Nonprofits need clear AI usage policies to prevent data privacy breaches and ensure ethical implementation.
[00:27:05] - Many software tools nonprofits already use have AI built in—leaders must understand how those systems handle data.
A Case Study: AI in Action for Donor Engagement
[00:22:00] - Example of a Florida-based humane society using AI-driven donor segmentation and messaging to reactivate lapsed donors.
[00:25:05] - A multi-channel AI-powered campaign increased response rates and donor retention significantly.
Future Trends in AI and Nonprofit Work
[00:27:56] - AI literacy will become a required skill for nonprofit professionals—just like using Excel or CRM systems.
[00:31:00] - AI’s role in streamlining donor engagement and making data-driven decisions will continue to expand.
Final Advice for Nonprofit Leaders
[00:33:02] - Nonprofit leaders need not fear AI but instead explore its potential by asking AI tools themselves how to start.
[00:33:41] - Approach AI with curiosity—lean into it as a tool for expanding social impact
GUEST BIO:
Josh Hirsch is an accomplished nonprofit strategist and educator specializing in digital communications and fundraising innovation. With over two decades of experience, Josh has driven impact for organizations through roles such as Director of Mission and Communication at Susan G. Komen Florida and faculty member at The Fund Raising School. He currently serves as Education and Training Strategist at Fundraise Up, where he leads the development of cutting-edge donor engagement initiatives, including the Modern Digital Fundraising Essentials certificate program. Josh is a generative AI expert, leveraging emerging technologies to optimize nonprofit workflows and enhance donor experiences. A dynamic speaker and consultant, Josh is passionate about empowering organizations to overcome barriers to giving and achieve sustainable growth. Connect with him at theaidude.ai or through his AI Dude podcast.
IMPORTANT LINKS & RESOURCES:
Soukup Strategic Solutions, Inc.
Indiana University Lilly Family School of Philanthropy
RELATED EPISODES:
-
Carol Hamilton: My guest today on Nonprofit Mission Impact is Josh Hirsch. Josh and I talk about using AI in fundraising and donor engagement. We explore:
The evolution of AI, and its potential to revolutionize how organizations connect with donors.
How to train AI tools effectively, leveraging data responsibly, and using AI to enhance—rather than replace—human-driven strategy.
Why nonprofit leaders who embrace AI thoughtfully do not need to fear the technology
I appreciated Josh’s description of the steps to train the AI model you are using – assuming that you are using one. I attended a workshop last year that described how to do this process. Since I had been using AI in relation to the podcast for a while, my model (I use ChatGPT) had quite a bit of my work already inputted. I walked through the steps described to create a description of my brand voice.
It was actually a humbling experience to read the description and in a very weird way feel seen. It described me as thoughtful and reflective, supportive and encouraging, educational and informative as well as practical and action oriented. This is what I strive for in my work. And to have that laid out in bullets was reaffirming.
This spring I have seen quite a number of posts of people having their AI create them as superheroes.
I tried it as well and I appreciated that it named me Strategia - the Clarity Catalyst and named my powers as
strategic sight – zooming out to see the whole system and zooming in to see root causes
equity field generator – neutralizing hierarchy induced silencing
facilita-force field – helps ground move from stuck to flow
evaluation alchemy – transforming meaning data into gold
and time bending retreats – where I slow time for deep reflection and speed it up to get shit done
I am armed with reflective pause – to stop time to help people think not just react
The reframe ray – to turn defensive yeah buts into curious what ifs
And the purpose pulse – to instantly reconnects teams to the heart of their mission
Rather than share the AI image of my action figure, I am planning to draw my own version. Yet I was pleasantly surprised that it felt accurate.
Maybe it is a little like reading a horoscope and feeling it resonate.
But if you go through this process, the model you use can learn your organization and its purpose, culture and voice – so that the first draft it gives you of any writing you are using it for is not so vanilla. It is much closer to you.
This is a foundational step to being able to apply AI effectively in some of the other ways that Josh and I talk about.
So I encourage you to give it a try and let me know how it goes. Did you learn something unexpected about your organization? Did it help you connect back to your why? Did you like what you saw in the mirror?
Welcome, Josh. Welcome to nonprofit mission impact.
Josh Hirsch: Thanks, Carol. I'm excited to be here.
Carol: So I like to start each conversation with what drew you to the work that you do? What would you say motivates you or what would you describe as your why?
Josh: So I'm going to go back to before I started grad school. And I had finished my undergraduate degree in advertising from the University of Florida. I was working for, you know, I don't want to say I wasn't motivated after school to like, go join an agency or go to like a portfolio school or anything like that.
It was kind of lost, and it was working a no nothing job in town for a local copier, basically. It was a copy printing service. And after a year, I'm like, what am I doing with myself? I, I, I need to, I need to go back to school, I need to go get more education. So I looked through the entire graduate catalog, and found a program called Family, Youth, and Community Sciences.
And at the time, at least at, at University of Florida, there wasn't necessarily a master's in nonprofit management. There wasn't, you know, focus on social impact from a degree perspective, but this program, when I was reading through it and then met with the graduate admissions professor, it's like, yeah, this, this is in line with what I want to do.
I want to give back. I want to, you know, I, at the time, I had the mindset of I wanted to work with youth, I wanted to work with kids from an informal education standpoint. So fast forward, finish my master's degree, and I'm like, well, if I'm here, I might as well get the master's of science degree and write a thesis.
Contextual features of animated violence leading to the learning aggressive
Carol: the,
Josh: basically is like,
Carol: Okay.
Josh: I said, yes I would, not knowing anything about fundraising. But I'm like, alright, this is a good way to continue to expand my career in nonprofits.
Carol: So, I'm going
Josh: first thing I did after, after leaving that, I was the director of development and training, so I had
Carol: good
Josh: You know, education component behind it, which is something I've always been passionate about
Carol: we
Josh: and went out and bought the first nonprofit book I ever bought. I think I still have it on my shelf, grant writing for dummies
Carol: company.
Josh: and just started immersing myself in everything I could nonprofit professional development.
I went to my local children's services council
Carol: sent it
Josh: attended some training there.
Carol: we
Josh: forward 15 years, having raised millions of dollars and gotten to work for You know, major national NGOs that are providing incredible services for women and men all over the country transition more into the digital side of fundraising and a consultant and then sort of working on faculty for university.
And that's what brings us to today, where I'm an education and training strategist for fundraisers where I get to develop thought leader content and ongoing professional development for. Best practices in the sector, not focusing on our platform. So, you know, platform agnostic really ensuring that all nonprofits have
Carol: States. A
Josh: education possible, from focusing mainly on digital fundraising,
Carol: do
Josh: that they can ensure greater social impact.
Carol: has
Josh: three minute version of how I am, where I am, and who I am today. I
Carol: Well, thanks. I, I love when people start kind of at the beginning and, and we've that, and it all makes sense in hindsight, right?
But when you're at that last point in your career, it takes a little while to kind of figure it all out. But I appreciate that. So within the kind of digital space one of the things that you dial in on is, the whole realm of AI and you've got a podcast where you help nonprofits kind of navigate that world.
And so I, I looked up and I, it was amazing to me that that chat GPT actually came out a couple of years ago now. And, and, you know, every software you open up now has some, I was reminded of the old Microsoft paperclip. And I was like, well, they're slightly more useful but sometimes they're not.
But anyway, how are you seeing nonprofit leaders really integrate AI into the work that they're doing?
Josh: Clippy, our First AI assistant before we really even knew what generative AI was. And you're right. Chat GPT has only been here since November 22nd, 2022. That's when the beta program went live. I was an early, early beta user. And then a week later on November 30th, it became mass adoption. And now chat GPT is just part of our lexicon.
Like us, we can't think about anything when we associate generative AI other than chat GPT. And that's great. You know, it's the. brand leader, the Xerox, the Kleenex, the Tylenol, they're the ones from a brand marketing standpoint have the, have it covered. But there are so many other tools out there and there's different purposes for each tool.
There's different focuses. But at the end of the day, if you understand two key components, In generative AI, you could be successful in whatever tool you're using. That's in training and prompt engineering. And when I say training, it's not about going to a workshop and listening to a speaker like myself, you know, teach you about something.
It's about how you train the model. And then from a prompt engineering standpoint, it's how you're actually talking to the computer. It is not about codes. It's not about
Carol: be
Josh: is about having just clear, linear thought explaining to,
Carol: with.
Josh: the tool, the whatever you're working with
Carol: So
Josh: output is.
So when looking at how the nonprofit sector and NGO leaders as a whole benefit from this, it goes well beyond the, hey, I need to draft an end of year appeal letter for my major donor segment.
Carol: to
Josh: And you're going to get something pretty vanilla. It's not going to be specific, not being your brand voice.
So over time, learning how to train the model, learning how to better prompt it, you're going to get something that's more in line with you. But going beyond just that one piece, think about it from a campaign perspective,
Carol: to
Josh: and I want to use the word fighting for, but we're all competing.
We're all trying to get after our donors' attention. When we know that
Carol: going
Josh: supporting multiple organizations. At the end of year or at all times
Carol: a
Josh: That donor is going to get three or four appeal letters from the, you know, different organizations they support. It's going to have mail merge variable data of address, name, giving amount, maybe it'll have some recommended giving, you know.
Pretty blah. It doesn't really speak to who I am. Yes, maybe you're gonna segment me off to your major donor you know list, so it's gonna be a little customized from that. But it doesn't talk about my intent, my interest, my connection to your organization.
That's where AI comes into play. Using clean data, using donor anonymized data, and when I say donor anonymized, I'm gonna pull out My name, I'm going to pull out my mailing address, and in an external spreadsheet, I'll know that
Carol: not,
Josh: 374 equates to Josh. I'm going to leave the zip code there because it's good for geolocation trends.
General quantitative data and demographics is good, but pulling in that qualitative data.
Carol: yeah. Right.
Josh: only generate one letter, but a true bespoke donor journey, speaking to who I am, keeping me connected is going to.
Take it. It's gonna take more time. Like, let's be honest, doing a traditional mail merge versus doing this focus on donor stewardship, retention and donor journey.
Carol: what
Josh: lifetime value of a donor is going to be so
Carol: mean
Josh: Then it is for the other organizations because you're listening to the donor.
You're talking to the donor. Can you do this for all 10, 000 people in your CRM? You could, but you really don't have the time to do it.
Carol: I
Josh: from segmentation purposes. You know, I always believe
Carol: you
Josh: There's two main areas we want to focus on, right? Those that are the most invested. So that's our major donors, our monthly donors.
Carol: to,
Josh: We know that we need to be communicating with them differently. Our monthly donors need to be getting consistent touches from us on a monthly basis.
Carol: team,
Josh: And it's not just, thanks for your gift, Carol. You've been able to do so much these last, you know,
Carol: the
Josh: days.
Carol: all the
Josh: Once again, using AI, I can now create a complete 12 month communications plan just for Carol, my monthly donor, who is very interested in our after school arts program.
But I can then go to donor Josh, who's also a monthly donor, who is more concerned about our after school STEM program, and focusing specifically on that. So taking out my monthly donors, taking out my major donors, focusing on, you know, my top ten percent. But at the same time, too, using both generative and predictive AI, it's like, oh, Carol's engagement score is going down.
She has the potential to churn. So AI can help me identify, it's like, wow, Carol's been a really engaged donor or constituent for the last ten years, but all of a sudden she's slipped. Let me reactivate a re engagement campaign before you completely lapse. So those are the two ways that I could look at it.
So, going well beyond the hey, draft a social media post, draft an email, but thinking about the donor, thinking about how we can engage them better, and looking at it from a campaign strategy perspective.
Carol: Yeah, and I wanted to follow up on a couple of different things he said there first. You know, regardless of AI, you need to have that information of what I'm interested in, which is a whole different process of making sure that you're asking and, and capturing that information so that you can do that more personalized approach. But then the other thing you talked about was training the model and, and, you know, obviously Folks do need to get training and how to use all these tools. But this is a different way of talking about it. Can you, can you say a little bit more about how folks can go about doing that and, and essentially make the model more personalized to them and their organization, I guess.
Josh: Absolutely. So you have three levels of training. You have zero shot, one shot and few shots are also known as multi shot. Zero shot is sitting down and saying, Hey, chat GPT draft me an end of your appeal letter for my afterschool arts program. You're going to get a letter. So be written for an afterschool arts program focusing on an end of year, but that's where it ends.
It's going to be very vanilla. It's not going to be specific to who you are. It's not going to touch on the impact that you're making in your community. And it's not going to be in your brand voice going to that next level of one shot learning or or, you know, from a training perspective, you're going to provide the model.
And when I say model, you know, tools like chat, GPT, Claude perplexity, Gemini, whatever you're using. So as you're training the model on your organization and on the campaign, you're going to provide the organization name, you're going to provide the U. R. L. Because in the L. L. M. Or that large language model, which is thought about as a giant database.
There's going to be Sunshine Academy, but there may also be 10, 000 other Sunshine Academies in the U. S. that are operating.
Carol: And
Josh: I'm going to include the name, the URL, I'm going to include some minor background information on the organization. Literally going on the website, grabbing the about us, the mission, some programs, and feeding that into the model. It's going to move further along and it's going to become
Carol: American
Josh: with who you are, but what we're really aiming to get to
Carol: I'm
Josh: few shot or multi shot level of a training.
That's where I'm taking past appeal letters, blog posts, social media, our brand style guide if we have one,
Carol: don't
Josh: understand what is my organizational voice, what is my organizational tone, based upon audiences, because we know different audiences need different tones of voices, specifically focusing on them.
And at that point, you've got a virtual you. You have an expert that understands your organization, that thinks and speaks like you. Now, to that point, we are never directly copying and pasting the output that we get from our first iteration of you know, working on a prompt and saying, we're done, sending out to our constituents.
There's going to be,
Carol: to
Josh: percent of the time, further revisions. Further, you know, structure refinement or looking at different areas you want to focus on. And even then too, after you've gone through 3, different outputs, I'm going to take that content, bring it into Google Docs, bring it into Microsoft Word, Further edit, further ensure that it is in my actual organizational voice and tone.
And at that point, I'm then going to say, alright, now we're ready to send this appeal letter, email, content strategy, program, whatever we're working on, out to our constituents.
Carol: Yeah, recently I was looking to do a solo episode about a topic and so I asked it for an outline rather than asking it for the full you know, full iteration and then I was able to just kind of walk through that and, and, and really have it sound like me. So that's a, you know, maybe a different way. Of doing that. But yeah, I think it can be easy to kind of fall into a well, it's done now and just grab it. And then you look at it again later. And you're like, oops, that doesn't really sound like me. Even with some of the training that you're talking about. So you're talking about training and prompt engineering.
What is it? Now. and we haven't really talked about Prompt Engineering, but what would you say is important for non profit leaders to really understand and know about AI as they think about using it to get within their work?
Josh: So, don't be scared of it, right? This is not going to come along and replace employees. What it is going to do, is going to allow those who are utilizing it effectively, to essentially have superpowers that they didn't realize that they had before. It's going to allow you to make better strategic informed decisions based upon data analysis.
It's going to allow you to create those personalized experiences for your donors. Now, do I think AI is going to take jobs? Like I said, no. However, the nonprofit leader who's utilizing AI effectively. is going to take their jobs because they are able to do more with less. They're able to create those better experiences, like I said, for their constituents and it's going to give them a deeper insight into your organization.
So we now have employees who are more focused, who can do more with less and it can create those, like I said, better experiences for your donors because they're using the data to help really drive everything overall.
Carol: So it sounds like and what I've, what I've heard and talked with other people about in this realm, it's, well, one, you're, you're, you're not just, it's not just replacing or helping you do kind of the, the, the more mundane things, it's actually taking those mundane things and taking it one step further if you've got everything lined up correctly but then I would think that it also allows folks to Spend less time on those those kinds of things and kind of move up in terms of their skills and really be using those more You know higher level more strategic if you're you know, if you're doing if you're using it well
Josh: Yeah, exactly. And using it well is key. You know, there's good prompting, there's bad prompting, and it takes time. Like, there's no magic bullet. What I think is very important is to think linearly, think clearly, and think both strategically and from a creative thinking standpoint. How do I want to achieve what my end goal is?
You know, so it's called root cause analysis. I know where I want to end up. I know where I'm starting, so I will reverse engineer and think backwards. And how can I utilize generative AI to help achieve what I'm trying to accomplish?
Carol: So, what would you say are some of the things that people need to think about when they're when they're creating those prompts What makes a
Josh: So, for, right, so being as clear and as specific as possible with what your desired output is. You know, some of my prompts will be 3 sentences long. And, and, and other times too, I will have different desired outputs within that prompt. Meaning, saying like, okay, accomplish this, and once you've done that, this is what I want you to do.
And then based upon this, do this next step. It's about knowing that you're not going to break anything, right? So like I said, don't be scared. Go into it understanding that It will not always be right the first time, and it has the potential for bias. Like, inherently, AI is biased based upon the data collected from the LLM, based upon our own initial bias.
So, identifying ways to create those feedback loops within whatever tool you're using, Chat GPT, Gemini as such, and learning from it just as much as you are just generating straight output. So I love prompting it and asking it, you know, things like, Do you understand or have any question? By prompting it with that, I'm able to get a better understanding if it needs clarity, if it needs more information.
At the same time too I will, like, let's say I'm working on a list of subject lines. I love at the end to say, And what's your reasoning? So I'm seeing, once again, perspective and thought. To get that, it's like, wow, I didn't think about it that way. Let me take my approach, possibly based upon its ideas.
We need to understand that this is a supplement and it's not a replacement. That
Carol: on
Josh: the clear and more defined that we can be with our desired output. The better quality content we're going to get.
Carol: And I heard of someone talking about it in terms of, you know, thinking of AI as your intern. And so you have to be, you know, very specific, very kind of one step at a time In terms of how you're communicating and I, I definitely have run into frustrations where I thought I was being clear and clearly I wasn't because what it spit out had nothing to do with what, what I was trying to do, what are, what are some mistakes that you see people making and as they try to apply it to their work?
Josh: So I think a big mistake is people jump in without thinking about the data privacy and accountability perspectives that we're dealing with. So we need to ensure, and this is like one of the biggest soapboxes that I'll jump on, is that you must have an AI usage policy for your organization. If you are using generative AI in some capacity, and like you said before, it's built into so many of our third party vendor tools.
that we're already using, that we may not realize that generative AI is happening in the background. So, you know, understanding their terms of service and how they're using data, but without having those guardrails in place within your organization from a policy standpoint, you've got people operating in silos.
You have people who may be using donor private information and not realizing what they're doing is not okay. At the same time, too, you've got a gift acceptance policy. You have a data usage policy. All these things help structure and provide guidance for your organization. That is why you need to have an AI usage policy along with these guardrails for your organization.
Carol: And what are some of the trends that you're seeing kind of emerging, you know, we've talked about it being around for about two years, how people are starting to use it. But what are some trends that we need to be paying attention to?
Josh: So I think we're going to continue to see such fast paced development in the world of generative AI almost daily. There's a new third party tool that's coming out. I think what's important is that I subscribe to like three or four different newsletters, stay informed, right? We got to know if this is something that's going to be a part of our lives, which it is.
You know, they're, they're starting to teach it at younger ages. I believe the next generation of workforce will, you know, it'll almost be like on a resume. It's like, well, do you understand how to use chat GPT, generative AI, whatever it is, because it's going to be expected from employers that, you know what?
You're gonna work 40 hours, but I expect you to put in 60 hours worth of work because you have tools like chat GPT that can help you do more with less. And there's nothing wrong with that, right? I think having tools available to us to optimize our time as such is going to make us better employable.
It's gonna make us smarter, and it's gonna make us have, like I said before, more superpowers than we had previously and allow us to expand our knowledge base, allow us to expand our skill set. And create greater impact within our organizations.
Carol: And are some of the characteristics of a successful implementation of AI in an organization? Can you give me an example of where you've seen a group do it really well?
Josh: Sure. So I'll use an example of a project I worked on for a client. So they're a humane shelter based in Southwest Florida. As you can imagine, summertime is hot. In Florida, and animals need to be rescued, they can't be on the streets, you know, we need to ensure that they find their forever home.
At the same time too, summertime is a lull in fundraising, especially when you have some potential for seasonal audiences based upon where you are, and there are a lot of, you know, as you call it, snowbirds in Florida. And summertime, if you're from the North, you don't want to be in, you know, down here, like, you want to go back home where it's cooler.
So, we identified the need to launch a campaign. And what we did is we used generative AI to build a persona of the Vice President of Advancement and Marketing for this Humane Shelter. They were trained, and once again, going back to a training component, trained on the brand style guide, trained on past appeals, impact reports, emails, social posts.
So it understood and took on that voice and persona of the organization. And we spent several months training this to get to the point. We had done some tests with it and in minor content before you rolled out the summertime campaign in July. At the same time too, we wanted to focus on LYBUNTS and SYBUNTS.
Last year but not this year donor and some year but not this year donors. Now we were limited in tech perspectives. So their CRM and I won't mention the name because I'm, I'm not a fan of it. I think it, you know, there are a lot of better other ones out there. They had 70, 000 records. Okay, great, you've got 70, 000 records, but a lot of dirty data.
And at the same time too, the CRM didn't talk to their email marketing software. So you had inconsistencies with, you know, let's say 10, 000 records in your email marketing, but you've got 70, 000 records in the other. And even with duplicate records, like, you had a large gap that was missing. At the same time, too, because the email marketing wasn't talking to the CRM, you weren't able to get a complete donor profile.
You weren't able to see engagement levels of your constituents. So there are a lot of issues there. So we couldn't segment from an email perspective. Live in silence. Unfortunately, I missed an opportunity. But from a direct mail component, we were able to. So we went into the database and we identified live on some sideburns going back to 2018.
They had a very engaged donor base leading up to the pandemic. And then, as you can imagine, things tapered off in 2020. 2021 time frame. So we identified that as our key. We had a segmented direct mail piece that went out to like I said, about 3, 500 donors. We also had 6 emails, 6 blog posts, and 6 social posts that went out across the time of the campaign in July.
This content was generated using that generative AI persona, so it was from the perspective of actually a rescued dog and cat from the shelter. And we told the story through this campaign in a multi channel sense, over these six touches. The segmented piece from the direct mail Completely blew me away when we saw the results.
So I was in a wrap up meeting with him, and he's like, one second, Josh. Reaches into the drawer behind him, pulls out a stack of response envelopes. It's like, wait a minute. People still have checks? They actually still send things back in? We had almost a 4 percent response rate on our direct mail, which was, as you know, in the world of direct mail, like, huge.
We had donors that were giving sometimes two or three times as much as they had given before and they hadn't given in three or four years. So being able to speak to them specifically the way that they wanted to be spoken to in a messaging format that resonated with them blew our goals away. So their initial goal was 25, 000.
We raised almost 40, 000 to have an engaged audience. Which was key. So now it's not about just lumping them back in with all your other communications, saying here's your newsletter, you know, as we always do. Now you have an engaged audience and we need to speak to them. from that component, too. So you have your, you know, fundraising campaign, but now I have a dedicated stewardship campaign just for these donors
Carol: a million.
Josh: was able to do, once again, more with less, utilizing generative AI.
Carol: So as we wrap up here on each episode, I'd like to ask guests, what permission slip or invitation would they give to nonprofit leaders? And invite them to consider it to avoid being a martyr to the cause. And as they work towards cultivating a healthier organizational culture. So what permission slip or invitation would you give to nonprofit leaders?
Josh: Love the question. So I would, once again, tie it back to generative AI and know that, as I said before, you're not going to break it. Literally sit down at a prompt in whatever tool you're using and say, Hey, Chachi, PT, Gemini, Co Pilot, I'm in this role for this organization. And give a little background on the organization.
We're interested in using AI, but I have no idea where to begin. Be humble, right? Like, listen. We know that this is new technology, but we know that It's going to be driven by leadership. So by going into the prompt and saying, we don't know where to begin using it, give us some recommendations. Learning from AI how to use AI.
As I said, don't be scared. Lean into it and know that it's going to truly create those opportunities to create greater social impact for your organization as a whole. It's
Carol: All right, well, thank you so much.
Josh: been a pleasure, Carol. I really had. I'm glad we had a chance to connect today.
LISTEN + SUBSCRIBE
APPLE PODCASTS ● SPOTIFY