EP 26: Using the Power of AI to Transform Your Fundraising Efforts

Jeff: Welcome back to the Elevate Your Event podcast, where we talk about all the many different ways that you can make your fundraising events better. Today in the studio, we have a special guest. We've got Sean Olds from Boodle AI. Is that right? Boodle AI. All right. Sean, go ahead and introduce yourself. Give us a little background and tell us what Boodle does.

Sean: Sure. Sean Olds, co-founder and co-CEO of Boodle AI. I got my start in the military. I went to West Point. I was a computer science major. I won't take credit for our tech stack. The programs I programmed in, they now teach in the history of computers class.

Jeff: Yeah, me too. I get it.

Sean: So I have a wonderful tech team and development team. But I did my time in the military, got out, went into building companies, and got the startup bug in the late 90s. Continued that until September 11th, went back into government service, did counterterrorism work with the government, mostly Southwest Asia and Africa, and then came back into building companies, both domestically and overseas. During that time, I also spent about 25 years building or serving on boards of nonprofits. I jokingly say the genesis of Boodle -- my co-founder has a very similar path -- is that we were both lazy. We both got tired of beating our heads against the wall trying to raise money for nonprofits, and wanted to find a way to make it more effective and more efficient. So we wanted to develop a technology, specifically data science and machine learning, in an easily consumable manner for nonprofits. I usually give this spiel on a Zoom, which we've all been on for three years. I equate it to Zoom. Most people have been on Zoom through the pandemic and using it more and more even afterwards. I tell people all the time, I'll take a bet with you. I'll bet that 90% of the people you've spoken to on Zoom could not explain to you how video over IP works.

Jeff: Yeah.

Sean: And the reality is Zoom has made it press-of-a-button simple. People don't need to learn that. We wanted to make machine learning the same thing. So that's what we did, and in the course of building our tech stack, we developed a few unique capabilities that we realized could be used by the commercial world as well. So we've worked with both commercial and nonprofit clients to help them understand better how people consume and donate.

Jeff: That's awesome. So it might be no surprise to our audience, we're going to talk about AI today, artificial intelligence. We're going to talk a little bit about it in general, but also about how it can be used to improve an event. Since this is the Elevate Your Event podcast, we'll make those connections. It's a fascinating background. Not too different from ours. I think we have some synergies as it relates to trying to figure out better ways to raise money for nonprofits. First, I want to thank you for your service, obviously, to our country. And then also thank you for your service to the nonprofits you've been helping to raise money for. I know that can be a labor as well. Labor of love, right?

Sean: Yeah, that's right. That's what we like to say.

Jeff: Anyway, fabulous that you're in the studio with us today because AI is one of those things that I think freaks a lot of us out, right? You've probably heard in the news, there's a lot going on. People are starting to say, wait, that's creepy and cool all at the same time. And it probably is a little creepy and cool all at the same time, what you're going to be able to figure out. Some of the data that we've analyzed on the nonprofit side, the tools aren't necessarily new, right? So explain, when we talk about the wealth engines of the world or some of those other tools out there, where I can give it a list of names and addresses and it comes back and gives me wealth information -- where does it go from there? What does AI do above and beyond that?

Sean: Yeah. And I'm going to address the first thing you talked about, the creepy part. Because we've spent six years dealing with nonprofit people who thought, oh no, AI is going to take my job. We're actually thinking about a new tagline: "Yes, AI is coming for your job, but only the boring parts." We've always advocated that the best team is the human-machine team, because at the end of the day, AI can do in hours what might take a human weeks, if not months, of analytics work. But what the AI can't do is what we're really good at as humans, which is the fundraising portion, the ability to engage. If you leave a machine to do a major gift fundraise, and it identifies Mrs. Jones as a great major gift prospect, and it makes that phone call, it's going to ask Mrs. Jones how she's doing. And when she goes, "Oh, I've got to get out of here, my husband just had a car accident, our son's got COVID, my daughter Julie just broke her leg, and I've got this board deck I need to prepare" -- the machine's going to say, "Okay, thank you very much, Mrs. Jones. Would you like to make a $5,000 donation today?" We as humans have the empathy to say, "Wow, is there anything we can do to be helpful right now?" That development director is going to make a note in their calendar to call back a week from now and check in with Mrs. Jones, maybe a month from now to make sure everything's good. And then finally two months from now, when the household is settled, to make that ask.

Jeff: Yep.

Sean: And so that human-machine team can perform at a much higher rate than either the human or the machine could do alone. As far as the data, I love your example of "I've got all this data." That's what the wealth screening tools and a lot of others do -- they just sell you data. I've watched nonprofits, I've sat on boards as I said for 25 years, waste so much money to be able to say, "We've got all this data on our donors now." Good. What are you doing with it? "Well, we don't know." What the AI does is it helps provide insights. You can build models. You can start to understand who individuals are. You can start to predict the very nature upon which they're going to act. It's not always a sure thing that just because somebody's done something once, they're definitely going to do it again. I'm sure you're familiar with a statistic that's been circulating in the nonprofit space for about the past decade -- 80% of first-time donors don't come back.

Jeff: Right, right.

Sean: So you want to look at the 20% who do, and that's what we did at Boodle when we started seven years ago. We looked at the 20% who do come back, and they come back because they have an affinity for the cause. What AI allows you to do is take literally billions of data points, combine them all together to start to understand a person's affinity. And that's not something an individual development director is going to have the time or necessarily the ability to do.

Jeff: I think it's not what the wealth information is going to give you either, right? Wealth screening data can tell you who's rich -- it may not tell you who's generous. It's also not necessarily going to tell you, to your point, who has an affinity for your cause, because it's not providing you that behavioral data to say, hey look, they've donated, or people like these people have donated in similar industries or types of nonprofits, right? I think it's what you're getting at.

Sean: Absolutely.

Jeff: Yeah, that's powerful. I think that can definitely inform the people who are getting on the phone and talking to donors. And I love your example about the fact that machines are good at figuring out the information, but they don't have the human concept of empathy built into them. At least not yet. And that's going to scare me when they do. It's funny, my autistic daughter is very similar, right? I kind of equate it to that. She's not able to process those types of things either. She wants very factual, I would say, surface-level conversations. And those don't work great for fundraising because fundraising is relational.

Sean: Correct. Yeah.

Jeff: So that's awesome. Anyway, so we take artificial intelligence and say, okay, so now it's going to give us some relational information. Let's talk about some examples of where we see that being used to improve fundraising, specifically events. Who's coming, right? Have you worked with different nonprofits on helping them understand not just who the rich people are that are coming, but where do you take it from there to help them?

Sean: One of the greatest pleasures as an entrepreneur is not necessarily the exit, the sale, or anything else. It's when your client teaches you how to use your own technology. We actually had a client who called us up all excited and said, "Hey, you know that interest list you build?" Part of what we're able to do is take in a set of people and then attribute to them their interests. And sometimes people get really surprised at the interests that come back. So this particular group had brought in the 550 people who had RSVP'd to their gala. They took the interest list, and then they used that to curate their auction items. They'd had 10 years of auctions, and they saw over a 30% increase in revenue over any previous auction because people were bidding on things they wanted, not just out of obligation.

Jeff: Right. That sounds great. And I think that benefits anybody. I don't care how long you've been running an auction -- sometimes your audience is probably fairly consistent over the years, close to you. Some of them might bid on anything you put out there just to support you. But there's probably some portion of your audience that's changing year over year. And I think it's fascinating to say, am I keeping up with that?

Sean: Absolutely. Because we tend to get stuck in certain ruts when we're running auctions. We have that donation relationship with a certain business, we know they're going to donate to us every year, and maybe over the years we're starting to notice that's becoming less and less interesting.

Jeff: I've been to events where you kind of see it. And look, I'm sure the computer would have figured this out way faster than me, but I show up at this event and I'm looking at the items that are getting no bids. Then I look at the audience, and it makes perfect sense, right? The audience consists of a lot of older retired people. All the items getting bid on were trips and wine. The items that were not getting bid on were the Children's Discovery Museum and the zoo, right? So unless they're taking grandkids, that's just not a good fit for who they are.

Sean: Correct. And this could have preempted that and said, hey look, this is not only what their interests are, but can you break it down into their spending capacities, or what parts of the event they would participate in? Can we know this person is a live auction type, but that one is going to be more of a silent auction kind of person?

Jeff: Absolutely.

Sean: So AI can be used ahead of time, it can be used at the event, and it can be used post-event. At the event, very much so -- let's identify people who are likely to bid on a live auction, let's identify people who are likely to bid on silent auction items. And even within a large silent auction selection, who's likely to bid on the sports stuff versus the wine stuff versus the kids' stuff? Now what you can do is targeted messaging. The development team at that point can say, let's make sure that Jane Smith knows about the sports tickets that are out there. She's a big hockey fan, hockey is a big interest of hers. So let's make sure she knows that. Instead of just mass blasting "here's everything in our silent auction," you can start to target certain individuals, making sure they're seeing what's going to be of interest to them and draw them into the silent auction.

Jeff: I think it's brilliant. I'm thinking about how we handle broadcast messaging inside of Handbid and saying we could take this to the next level with what you're describing. I'm getting into this engineering ideation mode right now. I could certainly see in the future saying, here's a category that we have, give me a list of people that have an affinity for that category, and I'm going to send them a custom broadcast introducing items in that category. I love that, and I can see engineering-wise how that could actually work. So I think that's really cool. This is the creepy and cool part. We're talking more about the cool part, but when you start getting broadcast messages sent to you, I think it goes beyond creepy because for me, I don't want to hear about items that I have no interest in.

Sean: I agree.

Jeff: Right? It's the same question people raise about the ads now. Do you have to get permission to track people inside of these apps, inside of Apple? People say, "Oh, I don't want to be tracked." But do you? You might, because you're going to get ads no matter what. So do you want the ads that are actually interesting? Maybe you don't. I don't want any ads going to my wife that are interesting because I don't want her clicking on anything and buying that stuff. But for me, yeah, sure. Right. I get it. Well, that's awesome. Let's keep going down this path. One of the things I was playing around with -- I thought this was cool because we get people all the time who are working with us on their item descriptions. I was looking at an auction item that was up and I said, what could ChatGPT do with this item description? And this was actually awesome. It probably expanded this more than I would want, but I had an item description and it was like, "Spend the day exploring one of Denver's most popular neighborhoods with the Best of South Pearl package. Begin your day with a stop at Nixon's for a cup of coffee or fresh java. Spend the day browsing local stores, Gracie's Boutique, Common Threads, etc. Grab a classic wood-fire pizza for lunch at Kaos Pizzeria and finally drinks at Uno Mas. You get a $20 gift card from Nixon's, a $50 gift card from Gracie's, a $50 gift card from Kaos Pizza." So ChatGPT turns it into, "Embark on an unforgettable urban adventure through Denver's vibrant South Pearl neighborhood with our exclusive Best of South Pearl Package. Prepare to immerse yourself in a day of exploration, indulgence, and discovery as you delve into the heart of this captivating community." I'm like, this is good, right? And so this goes on. I'm not going to read the entire thing because it'll take up the entire podcast, but I love stuff like "sip on a steaming cup of perfection while you plan your itinerary for the day, knowing that your mug and a $20 gift card will keep you fueled and ready for an exciting experience." So I'm going through, I'm like, this is awesome. And then I'm thinking, okay, I can't condense this. I would love to put this in my mobile app for people to read because I'm already ready to bid on this item. But what can I put on a display sheet? So then I told ChatGPT, all right, I need the TLDR -- the "too long, didn't read" response. And it came back with: "Best of South Pearl Package offers an urban adventure in South Pearl neighborhood. Package includes coffee and a gift card from Nixon's, shopping opportunities at Gracie's Boutique with gift cards provided, lunch at Kaos Pizzeria is included with a gift card. The day ends with drinks at Uno Mas Taqueria and Cantina. This package provides a chance to indulge in unique experiences." Got it. I love it. So look, for people who dread writing item descriptions, this started with a very, I would say, okay-written but not greatly-written paragraph. And I got two versions of that in less than 30 seconds.

Sean: Generative AI is an amazing technology. It really is. I mean, you look at the adoption rate. The telephone took 75 years to get 100 million users. ChatGPT took two months to get 100 million users. It blew any other previous technology out of the water. The nonprofit space is known for not being quick at technology adoption. I was at the AFP conference in April. Out of, I don't know, 60 or 70 sessions, there was one on ChatGPT. I went in figuring it would be empty. There were 300 seats. It was standing room only. And whenever I give speeches on AI within the nonprofit community, I always start with, in the first 30 seconds: "How many of you used AI this week?" Before ChatGPT, less than half the hands in the room would go up. And I'd say, okay, for all of you who don't have your hands up, how many of you use Netflix? How many use Google Maps? People don't realize AI is in their life every single day. It's there, they just didn't realize it. At this conference, the speaker on the ChatGPT session asked a very similar question -- how many of you have used ChatGPT? In a room of people who are usually not quick tech adopters, over 90% of the hands went up.

Jeff: That's amazing.

Sean: ChatGPT, generative AI, is a very easy-to-use technology, and people can figure out how to get it into their workflows. Just your example -- "I've got to write 10 item descriptions. That's usually going to take me a few hours." Now you can have it done in about an hour, and it's probably going to do a better job than you would. Now, you still need the human-machine team. I don't know if you saw the news article last week -- the lawyer who will probably no longer be a lawyer much longer, who used ChatGPT to cite a bunch of cases that actually didn't exist and then submitted them in court. He didn't vet it. He just took exactly what ChatGPT gave him, took it as gospel, and put it out there.

Jeff: Yeah, and I've read some of the stuff it writes and you do have to correct it from time to time.

Sean: Absolutely.

Jeff: So I definitely understand that. What are some other things that nonprofits need to know about the risks of using ChatGPT?

Sean: Outside of making sure you don't just cut and paste, one of the other big things is you don't know where your data is. If you're putting data in, if you're putting any information in about your donors, about anything in your nonprofit, it's owned by the generative AI and they can do what they want. OpenAI made a statement after using people's data to train that they're not going to do it anymore. But there are, at last count, about 100 LLMs out there, and those assurances aren't out there with all of them yet.

Jeff: So just for our audience, can you define what an LLM is?

Sean: Sorry -- large language model. They've been around since about 2017, I think, was when the first one started in development. What makes them unique is you can't just decide "I'm going to start an LLM company" and have one up and running in a few months. These models take in vast amounts of data, all unstructured data, and are able to just assimilate it. And in doing so, be able to give very defined responses back to people. So that's why when you look at OpenAI, which ChatGPT is built upon, they'll talk about how many data points they used to build that out. I think the last one I saw launched was over a trillion. That's the new record. When you've got that much knowledge for it to pull from, it can produce amazing responses very, very quickly. But all of the LLMs right now do something called hallucinating, which we as humans sometimes do too, right? In fact, OpenAI last I saw states that over 30% of the responses will be hallucinations.

Jeff: Wow.

Sean: We've seen that. It'll give one response that is amazing, and then it'll give another response that sounds really convincing, but you do just a little bit of research and find out it's completely inaccurate. That's where the human-machine team comes in to check it, make sure it's good. But if it is, then run with it. And that's why, if you've got people who are -- this is why I say AI empowers people, right? If you're trying to turn a 21-year-old college graduate who's walking into your organization into your VP of marketing using ChatGPT, it's not going to work. But if your VP of marketing needs to do some quick stuff that might take several hours, they can now do it in probably half an hour. Your job is safe, Kristen.

Jeff: No, I get it. It definitely can speed things up. And I think it can also be a good starting point for some of the things that people want to do. I would still go back through these item descriptions. Even with the TLDR response, I'd throw back in the values. It did pull out the fact that it's a $50 gift card, and I think that's helpful. But I think it's a good basis to work from for item descriptions or maybe even donor emails and responses. And sometimes I'm just curious. Our staff uses Grammarly too, and Grammarly has an AI engine built into it now called Grammarly Go. It's funny because you could do something very simple, like write an email and then say, "make it better." I'd say 60% of the time it's better. Sometimes it just rewrites what you said. But what's funny is they have an option that says "write an apology email." And of course I'm thinking, okay, you have to give it what you want to apologize for. But it's fascinating sometimes what these things come up with because they definitely -- like, I like that sentence, I'm going to use that. I'm definitely not using the next one. In terms of donor correspondence, whether you're writing that invite email and trying to really incentivize people to come to your event -- I love those moments, right? I've written my invitation, and now I'm going to literally inside of the editor in Handbid hit Grammarly and say, "make this more compelling." I can tell it that. And then all of a sudden it's rewriting. I'm like, yes, I like this, right? But now take it a step further. Think about an event. Right now when you try and convince someone to come to your event, you're writing one generic letter that's going to go out to hundreds or thousands of people that you're trying to invite. We created Boodle GPT, which is a middle-layer AI company.

Sean: Sam Altman, who founded OpenAI, when he was talking to Reid Hoffman, mentioned that the next generation of companies are going to be middle-layer AI companies -- companies that have proprietary data, proprietary insights that they can sit on top of an LLM, that then allow them to give deeper insights into an industry or a vertical. So now imagine, because ChatGPT knows nothing about your donors or who you're writing to. You could feed it a prospect list of 1,000 people. You can feed those thousand names in. It doesn't know who those people are. With Boodle GPT, because of the identity resolution engine that we've built, we can identify who those people are in the real world. Then we can match them to the 35 billion insights we have. Now, with the speed of generative AI, you can ask it to write letters to invite people based on their individual interests. So all of a sudden, it can write a thousand letters for you in no time at all, but very specific to each individual's interest. If you are -- and you could feed it in, hey, by the way, these are my auction items. So now it can write towards that person's interest and highlight Jane's interest in hockey and Bob's interest in wine and be able to pique their interest even further rather than just one generic letter.

Jeff: That is cool. That is really cool. And I think what it does, to your point, is it's really going to drive a higher conversion if you're able to get people interested in what you're doing. Have you seen it used even in terms of how you would organize a mobile bidding interface? Have you thought about how that could work where, based on who you are logged in as, I could resort and show items in a certain order?

Sean: I would think if the engineering was done on the front end, it absolutely could be done that way, where this individual shares these interests, so we're going to pop these items up in front of them. Absolutely. You still give them the opportunity to look at the entire suite, but why not draw them in with what's going to be most interesting to them?

Jeff: I think even just sorting the categories that way.

Sean: Absolutely.

Jeff: Yep. I love it. This is so exciting. You know what I think we could use this for? Getting auction items.

Sean: There you go. True. You talk about the job that people don't want.

Jeff: And I think if the people listening are involved in building and doing silent auctions, they're probably remembering what Sean said at the beginning of this -- that AI is going to eliminate the parts of your job you don't want. That's probably part of the job they don't want.

Sean: They don't want. But I will tell you, another way they can use AI is in understanding the people who are there. If you can show that, hey, I'm bringing 550 people to my gala and they have a high interest in jewelry, now you can go to a jewelry provider in the area and show them, here are the people who are attending, here's their disposable income. Basically, I'm showing you your market. Would you like to donate an item and have a lot of advertising in front of your market? So now it's not just, "Hey, will you give to us because we're a charity and you have no idea who's coming to our event and then maybe buying your item at a huge discount?" It's "I'm putting you in front of your target market who will potentially come see you."

Jeff: Yeah. I love it. And I think you couple that with some AI assistance in how you actually deliver that message. Because some people are comfortable cold calling and getting on the phone and asking. And some people aren't. But take that person on your team -- because I think everybody has one, and you cling to that person on your auction team who just, I don't know what it is, but she gets everything, right? Because she's not afraid to ask. They're the ones always emailing saying, "I got a donation from this place, it's coming in the mail." I'm thinking, how do you get all this stuff? Take that person. Whether it's a series of emails going out or a series of phone calls that are AI-generated, how cool would that be? Because once you've gotten all that information down -- 550 people coming, they've got an interest in jewelry, specifically this type of jewelry -- and I think if you donated this type of item to our event, it would go a long way to exposing you to a great audience who will ultimately see you as philanthropic, be interested in your item, and might show up in your store.

Sean: Exactly.

Jeff: No, I love it. It's endless what you can do. I know it's overwhelming for a lot of folks listening. I think some things that you mentioned are important to reinforce. ChatGPT to me is fun, and it did a great job of showing my daughter what a five-paragraph, three-argument essay should look like on the symbols of Christophany in Lord of the Flies. She was really disappointed when ChatGPT came out with that in less than 20 seconds. She worked on it for a week. But it's cool with that stuff. When it comes to your donors' personal data, though, it's really important to be careful. Because the last thing anyone wants is that data being exposed, and these companies will own it. Microsoft will own it. So let's talk about Boodle for a second here. That middle-layer GPT that you're talking about -- you're working with companies like Handbid, integrating that technology directly into the software so that data stays safe inside of the application, right?

Sean: Correct. So Boodle as an entity has always operated as a data processor. We don't sell data. That's not our business model. Our terms of service forbid us from selling, sharing, or co-opting. Basically, we can't do anything with the client's data other than what they direct us to do. So in a lot of ways, we act like a CRM or a cloud storage entity. Because of the way we built, what we realized when we started interacting with generative AI is we could actually act as a safeguard for organizations. By building Boodle GPT, we can do everything with the data inside our application layer and simply use the generative AI to help generate the response that the user needs, but never actually putting any of their data into the foundational LLM we may be using. We may switch off of ChatGPT at some point, go to Bard or Anthropic or one of the other ones out there if they become better suited for the nonprofit space. But what's important is they never touch the PII. We keep it in our application layer under our terms of service, which strictly prohibits it being sold or anything else. And you're absolutely right. A lot of organizations don't realize that they're putting their own data at risk. We started from day one with the principle that we don't sell data, we sell insights. We consume billions of data points to build the insights that we have, but what we deliver are insights that you can then act on.

Jeff: Yeah, and I've seen some of the insights. You've shared them with me, and they're unbelievable.

Sean: Thank you.

Jeff: And I think what's amazing is that everybody asks, "So what would I do with this?" Well, what's nice is you couple that with a little bit of coaching from your provider or whoever to tell you how you can act on that data. And we don't have dissimilar conversations with folks who are always like, "Hey, do you integrate with Salesforce, or do you integrate with Bloomerang, or do you integrate with Blackbaud?" Okay, yes, but what are you going to do with that information? It's one thing to synchronize transactions over there. In another podcast, we've talked about what you can do with activity data. That activity data can inform future conversations and decisions, but let's keep that data inside of our own ecosystem and then let AI use that in the future to say this person has a propensity to go to these types of events, or not, right?

Sean: Yeah. And that was one of the big reasons we jumped into building Boodle GPT. To your point, we felt we were taking a big step. Instead of just saying this person is wealthy, we were saying, we've delivered the insights to show this person has an interest in your cause, this is the ask range based on how much disposable income they have, and that can inform what to ask for based on their philanthropic ability. But even with that, people might not know how to get there and understand it. With generative AI, it allows a development director to walk in, have a question, ask that question, and then be able to capitalize on our 35 billion proprietary insights and get the answer back that they want. "Who at my event is capable of donating $10,000?" Boom, here comes the list of names.

Jeff: That is so cool.

Sean: Or take it a step further: "I'm doing an event in the Dallas area. Here are the last 3,000 people to attend my event more than once. How many people in Dallas look like this? Build me a prospect list so that I can reach out digitally, email, phone, however I want to reach people." So we can actually help them build net-new donors in an area where they're doing an event, running events, other things like that.

Jeff: That gets me excited. I think about all the different fundraising possibilities you can do with that and obviously going well beyond just what auction items should I have, but let's get the right people there first. Now, the danger in that is we might find out that your wife likes wine and be able to invite her to an event, but sorry about that.

Sean: Yeah, we have ways in Handbid of limiting spending, which doesn't sound like a great fundraising tool, but we do commercial auctions as well and some of the commercial clients like that. But yeah, I've thrown that one at her before. Unfortunately, she knows more of the Handbid code than I do. So I think she would find some back door for herself, I'm sure.

Jeff: Fair enough. But this has been, I would say, enlightening for me, hopefully for the audience, just kind of what's coming on the horizon. It's arriving faster than any other technology I've seen in a while. Even as fast as mobile took over the internet, that technology still took 10, 12 years to really get full adoption. This is going to go much faster than that, I think.

Sean: It absolutely is, and the only thing I would say is there's no reason to be scared of it. It's now easier to use, as demonstrated by the adoption that's already happened. And the neat thing about the generative interface is that people can use it for everything from their job to settling a debate with their officemate about who was the highest scorer on the 1985 Celtics team. It does a wide variety of things. It doesn't just have to help you at work. It can help you in a variety of other areas.

Jeff: It does. The other day, it was funny. I was like, what can I get out of AI for cooking? I like to cook on the side. So I said, all right, I've got four Roma tomatoes, a yellow onion, a white onion -- I gave it a list of ingredients. And I said, give me an Asian recipe, an Italian recipe, and a Mexican recipe. And it did.

Sean: Yep. It's phenomenal. You would have spent hours on Google trying to find that.

Jeff: Yeah. Now, it didn't tell me which ones were the best. And it wasn't going to cook them for me.

Sean: That's next, I'm sure.

Jeff: There is this device. It's funny. I have a friend from Poland, and apparently these are really popular in Europe. I forget what they're called, but it's literally an entire machine that cooks everything. You just dump all the ingredients in, hit a button, and it cooks it. And I said, part of the enjoyment is cooking, right? If you're a writer, part of the enjoyment is writing. And if you're a salesperson or a fundraiser, part of the enjoyment is talking to people. Most people wouldn't want to give that part up, but obviously, use the technology in ways that make it easier. Well, thanks for coming on today.

Sean: Thank you so much for having us. And I'm sure we'll have you back.

Jeff: I'd love to have you on again in the future. We're talking about where things are going, and I'm sure we're going to get lots of questions from our audience. How do I leverage this technology? So if that's the case, reach out to us and we'll start connecting you with ways that you can leverage AI in fundraising and connect you with Boodle if we need to do that. And we'll start building more of these insights into our own platform. Obviously we're working on it. We'll wrap this podcast up. Thanks for listening to the Elevate Your Event podcast. If you like this podcast, please give us a five-star review and share it. Share with people that you think could benefit from the information we're sharing. Until then, happy fundraising.