Trusteeship Podcast Episode 49: Artificial Intelligence and Fundraising

Podcast

Aired: September 24, 2024

Artificial intelligence plays a key role in digital transformation, allowing institutions to scale their efforts and streamline their work. In this podcast, David Bass, AGB’s executive director of philanthropic governance, and Fundmetric CEO Mark Hobbs discuss the use of AI in fundraising in higher education. Effective use of AI and an improved data infrastructure can help you prioritize and personalize outreach, adapt to changing donor demographics, and identify untapped donor potential.

Transcript

Introduction:
Welcome to the Trusteeship podcast from AGB, the Association of Governing Boards of Universities and Colleges. We cover everything higher education leaders need to know about the challenges facing our nation’s colleges and universities. More important, we provide the facts and insight you need to solve those challenges and to be the storytellers and advocates higher education needs.

Artificial intelligence plays a key role in digital transformation, allowing institutions to scale their efforts and streamline their work. In this podcast, AGB’s Executive Director of Philanthropic Governance, David Bass, and Fundmetric’s CEO, Mark Hobbs, discuss the use of AI in fundraising in higher education. AI and data infrastructure can prioritize and personalize outreach as well as identify untapped donor potential to adapt to changing donor demographics.

David Bass:
Hello, I’m David Bass, AGB’s, executive director of philanthropic governance. I’m delighted to be joined today by Mark Hobbs, CEO of Fundmetric. Fundmetric has been a leader in putting AI to work in fundraising, we’re going to explore that topic today. In a matter of a year or two, AI as a fundraising tool has gone from being an exciting idea to, I’d argue, a significant force multiplier and strategic advantage for those organizations that are leveraging it. Before we dive into how AI can be used to identify the next generation of major gift donors, I’d like to provide some context about how younger generations of alumni and donors differ from those who have been the primary drivers of the growth in giving to higher ed in the past few decades.

A lot of institutions are concerned about potential enrollment cliff resulting from regional declines in the traditional college-age population. There is, however, also a looming donor cliff. Institutions have come to rely on a steady stream of major gifts, which account for the vast majority of funds raised from individuals. A lion’s share of those gifts come from older donors, who are a shrinking proportion of alumni, and who have different relationships with our alma maters and different approaches to philanthropy. While Gen Z and millennials account for 48% of alumni, they make up just 22% of alumni donors and contribute just 7% of dollars. In contrast, silent generation and Boomer alumni make up just 17% of living alumni but account for 57% of alumni donors, and 64% of the dollars contributed.

So why this difference? In part, younger alumni are more likely to feel burdened by student debt. Perhaps more importantly, they are also more likely to feel disconnected from their alma maters and report less satisfaction with their student and alumni experiences. That has a big impact on current and potential giving. Connected alumni are 23 times more likely to give than those who report being unconnected. These challenges are even more acute among minority alumni, who represent a growing proportion of alumni and are more likely to report concerns about dissatisfaction and disconnectedness.

This issue is going to be addressed in depth at AGB’s Foundation Leadership Forum in January in Washington, D.C. I hope those of you who are affiliated with foundations will be able to join us there. Mark will certainly be part of that program as well. To get us started, I’d like to think really about how AI fits into developing a sustainable business model. Sustainable business models and digital transformation are among the top strategic issues facing higher ed today. When Mark and I were talking before this call, he used the phrase “future proofing” to talk about the potential benefits of AI for advancement. So Mark, can you start us off by making the case for AI from the perspective of future?

Mark Hobbs:
Sure. David, and thank you very much for the invitation to be here today. When we think about future proofing, it might be important to start by stepping back from AI and just looking at future proofing business models in general, and the role that technology plays in that. AI being one of those technologies. But when we look at it, we look at it from a perspective of is the infrastructure, the systems that we’re going to build technology on in the future, in place? And a key piece of that is going to be the data set that you’ve built over time with individuals at scale.

ChatGPT may be the big technology today, there’s no guarantee that only ChatGPT is going to be required for tomorrow. So how do we build technology and the real infrastructure? And I believe it’s basically instead of building technology that can fish for people, teaching people how to fish with this technology. And really looking at, okay, as the market changes or as we have to engage young alumni on different platforms in different ways with different timing, do we have the ability to pivot, and how quickly? The way the data is organized can flip very quickly. And AI provides one of those methods that’s able to interpret things very quickly, that can spot changes in advance. So you’re using AI not only for the actual application of how you fundraise and how you prospect, but you’re also using it to look at what technology do we need to be building tomorrow, based on the changes we’re seeing in how people use it.

The final element is volume versus scale. We’ve seen business models that operate by volume, and scale is just putting an efficiency factor in there and saying, “We may not be reaching as many people but we’re reaching the right people.” And that allows smaller organizations to start at a different level, different pace, and still get that transformational impact. So in order to get to AI the underpinning piece is to have the data set, rather than just applying the technology itself. Which you can do, and if you have the data set and you have that ability to pivot to the various technologies that are going to come along, you’ll find yourself in a much better position with regards to future proofing.

David Bass:
One of the things I hear all the time from chief advancement officers leading some of the largest development programs is that even with hundreds of gift officers, a large number of their alumni with the capacity to give go unengaged and uncultivated. At the other end of the spectrum, community colleges and other small advancement shops that may have only have one or two gift officers still have thousands of potential prospects, they probably have much less carefully structured and comprehensive data to work with. I’d be curious if you could give us two snapshots. One, how some of the sophisticated, well-resourced advancement programs are using AI and two, how similar tools may or may not be accessible for more shoestring operations.

Mark Hobbs:
If we look at the well-resourced group first, I think what we’re seeing as a very effective use of AI is absolutely turning that model of gift officers having to reach out to thousands of people and not being able to do it on its head, and say, “Well, let’s start with the group who’s actually sending us a signal that they’re interested.” So a concrete example of that would be you can send out a personalized video through many different platforms, and you can in some of these platforms tell how long somebody is watching it. How much time and attention are they paying? If they are due for a gift that allows us to prioritize them much faster as they’ve recently engaged meaningfully, and take the people who are sending a signal, and also respect the people who are not. So the big organizations are able to use attention and engagement as a prioritization lens that isn’t there without this.

But on the small scale, it’s exactly the same thing. Because you only have a limited group of people, if you can tell who’s engaging, you can do that at scale. And it works to engage a larger group. As long as it’s personalized and as long as it’s relevant to your audience, they will want to receive that. The problem with volume is we’re not looking at whether it’s relevant, and we don’t often look at whether it’s actually engaging. I think there’s very easy ways to get data. I’ll tell you a story, for a small institution, they do opt-in for parents. So they ask parents upfront, do you want to receive these communications? And if so, how often? How many? And that leads to a much higher engagement rate of parents. The same thing will generally cascade with those younger donors. So by allowing yourself to focus, you can really get some leverage without a big investment upfront, and without having to do the same complex programming that you’d have to do across an organization.

So that really does allow anybody to access this technology, not necessarily need go all the way to AI, but be preparing their data and structuring it in a way. And all of that engagement data, whether it’s how long somebody watched a video, or filling in the opt-in form, is automatically captured. So because of that automation, it’s captured consistently. We’re not taking up resources with people manually entering it into a keyboard. So that’s a fundamental difference in the sort of business model that you’ll need in the future. What we do see is the people who are used to entering into the keyboard are now becoming the analysts. Because they naturally understand the problems that were previously in the data, and they can transform those into insights. That leads them to a different growth path, different set of education, but it’s a natural education that’s on the job. And so the insights are often really simple. They’re not spawning a pattern or an algorithm that’s very complicated for people to understand, but make a huge difference in terms of the output.

David Bass:
One of the things I think I’ve heard you say was one of the benefits that these technologies can have for smaller organizations is helping them build the data sets that will be the critical asset going forward. And I think that’s often a corner that gets cut at smaller organizations is investing in building and maintaining their data.

Mark Hobbs:
I think the other major thing is that there is a language that people who administer data use, and there are requirements that they often get requested when building a list. I think oftentimes some of the people who are asking for those requirements don’t necessarily know what’s in their data set. And some of the people who are pulling the data set don’t necessarily understand the business requirements needed to go out there and raise money. And so one of the things that a system like this can do is provide that layer of translation and really allow both parties to access data in different ways. And so you get this ownership of a new system, and that’s really key to the partnership is the joint ownership of it. So sometimes it’s the vendor provides the system, the limitation of the system is X. And when we certainly approach it, if we see a limitation in our own system and it’s not providing what works, the answer shouldn’t be, “Well, this is the way we do it, and so you have to adapt your system.”

And AI allows us to do that very rapidly. To have a code base that is able to pivot to business operations, as opposed to organizations changing the way they do everything in a CRM because that’s the way the CRM works. So there’s this meta-level to how this business model is evolving and it allows us to do things that I think would take a traditional software shop a lot longer, and that builds upon itself. So people begin to expect more of their software than they initially do with some of the legacy systems they may have. Weave those two things together and you get something that evolves with you, and that goes back to that future proofing is, we’re not just standing there and saying, “Well, this is version one.” There’s version X and it’s constantly evolving.

That’s the goal and that’s the new business model that you’re seeing able to enumerate, and that will very quickly cascade across most technologies. You’re seeing it at some of the major ones already, where they’re using some assistance to get things out faster. But what I’ve been amazed at is how the constituents respond to personalized communication that’s sent at the right time, according to their data. And the ability to have them opt in to these types of experiences, gives us the attention that I think is missing with some of the younger generation, because they’ve never been asked to participate in this way. They were just sent it, and we watched the conversion rates.

David Bass:
In the early days of AI—and by that I mean maybe the last year—I know a lot of advancement programs are looking to technology to support data mining and start capturing data on people’s levels of engagement, engagement scoring, some scoring around perception of the institution. What are some of the ways that AI is changing the ways institutions identify prospective major donors, and actually starts to cultivate those prospects?

Mark Hobbs:
At the highest level one of the best ways that it does is it adds a new perspective of behavioral data. Which is slightly different than the classic definition of engagement. There’s multiple types of data points that are important to AI, and they all do weight into what we see differently, but recency is really one of the things that can make a difference. So when we start reprioritizing portfolios we start to see more success, because we have more recent behavioral data and we also have a strong set of what this person doesn’t like. So we’re able to use those. And that’s sort of the non-generational nature of AI that’s changed I think in the last year. And now you can start using that data to build a stronger model. So if somebody says, “No, I don’t want to take a visit,” and they’re predicted to take a visit, that’s just as valuable from getting the machine to learn that it was incorrect as it is correct, from a data perspective. Ultimately you are going to reach that goal if you keep repeating and learning from it.

But judging the results of a model without recognizing that, yes, we need to be able to measure, but we also need to understand that a bad result is actually very good, because we can teach the machine what not to do and make it stronger. Another important point for us is that, from the fundamental perspective, is we try to do things that are not already being done. So we do not try and compete with the prospecting that you’ve already done. We think that the traditional methods play a key role and have worked and continue to work, but we know that there’s new ways to find diamonds in the rough, so to speak. So we will actually turn AI to find people who have never been screened and never been looked at. We will use some traditional well-screening information in the machine learning, but we won’t actually tell it how important a prospect researcher might think that data point is.

We will let the machine figure that out and then we will use the positive and negative results of that to make it even smarter. And eventually and very quickly, to be clear, it produces a new cohort of people, and then you can go inside and look at what was important. It rolls across the pipeline all the way to annual giving, in terms of automatically encoding things to establish interests that then are fed. So sometimes when I look at some of this data around Yale Alumni, it’s like you can see a pattern emerging with these people, but they’re just, they haven’t been on the planet long enough to accumulate the wealth. But you’re starting to see a certain segment of them actually value their education differently as they get older.

Certainly our data is showing you’ll start to see a larger chunk of those Millennials and Gen Z come into valuing their education, depending on where they are. There’s certainly a different conversation about change, but we do see in the data very clearly that people who don’t have that behavioral data aren’t collecting what people are doing both online and offline. Not just “Did you volunteer? But, “How meaningful was that?” And there’s different ways to measure that, but technology plays that key role in getting there, as opposed to manual recording. And so that makes a difference.

David Bass:
Worth bearing in mind too, that we are in the early stages of a pretty monumental intergenerational wealth transfer, and all the more reason to be paying attention to alumni who may be coming into very, very significant pools of wealth 5, 10 years down the road. You were also talking of course about how long it takes to cultivate a donor from first gift to major gift. That’s a long-term proposition. You’re planting trees for somebody else to sit in the shade of.

A couple of questions related to that. Have you seen any cases where institutions were able to really quantify the return on investment that they’re recognizing as a result of their use of AI and incorporation of AI in their larger advancement operations? And two, somebody’s going to have to pay for it. What’s the business case for AI, especially in a period when resources are short and institutions are getting more and more concerned about declines in tuition revenue?

Mark Hobbs:
Yeah, so I think we’ve spoken a lot about AI as an advancement model, that’s certainly where we focus. But you can apply the same approach, the same technology to things like modeling out degree completion. And then you can turn around and use that data and those results to inspire a large segment of donors who actually care about degree completion. I think there is millions of dollars in most databases that has not been discovered because of the current business model is if you want to raise more major gifts, you’ll hire more major donors. So it’s volume based, or it scales with people. I think scaling with technology is a much better investment. And I think as far as costs, the business case is simply what would it cost you to do this internally? Relative to the returns this is a no-brainer, in terms of what you’re going to have to put in and what you’re going to get out.

The bigger challenge is setting that agenda. So when you look at board chairs who have to contend with, “Well, we have all these different technology options and we have all these different approaches, how do I choose what’s right for me?” And my approach to that has always been, “Well, first kind of set your timer. Are you looking to make a big change in your time, or are you prepared to look out 30 years and say, I want to be making a change on that time horizon.” And based on wherever you can land with that, then you can come back and say, “Okay, this infrastructure investment is going to be set up in such a way that we’ve budgeted for the reoccurring maintenance to be able to evolve with it.” Those types of considerations make a huge difference in setting up the returns.

So I think the business case is simply that to do this, to get this technology, to get these results, to get the staff that you would need to do this is going to far outweigh the investment, even on an ongoing basis to maintain this infrastructure and that staff is going to be able to do more with less. You are going to be able to achieve this and the constituent experience when you start being able to personalize at scale because you have this data from a first party rather than third party.

So I always tell groups, don’t ask people what they’re interested in. Get them to tell you by their actions, right? And get them to tell you in ways that you are not guessing. We don’t say because somebody watched video for five minutes they’re more engaged, until we actually know that by having them confirm it. Technology allows you to do that, and I think that alone creates an ROI that’s going to be needed, because the rest of the world is headed that direction. And it’s important to remember that the brands are trying to get their attention just as much as you are. Being able to rise above that noise is a worthwhile investment alone. But you will see the financial, actual, tangible payoffs of planting these seeds early. Both with immediate dividends, more major gifts through the door, but in the next generation actually set up.

And right now we see a misalignment on the metrics a lot of the time for annual giving, when they’re actually planting the seeds for the next generation. But they may have metrics that measure participation, and that may be having them forgo that planting season that’s so critical. So being able to point that out and balance that and not abandon traditional methods, but rather work with you to evolve from them, makes a lot of sense. But it does require some investment. Not, I think the big investment people may think it is, but that’s my perspective and I work for an AI company, so take it for what it’s worth.

David Bass:
Well, thank you for that, Mark, and it’s been great talking with you today. I will look forward to seeing you at the Foundation Leadership Forum coming up at the end of January in Washington, D.C. We’ll be exploring this topic and a lot of related issues, not only the donor cliff, but a spectrum of concerns relating to fundraising, governance, endowment management, and other issues of concern to foundations there. So thanks very much for joining us today.

Speakers

David Bass

David Bass
Executive Director of Philanthropic Governance
AGB
David Bass provides thought leadership on board governance and best practices that relate to philanthropy in higher education. He oversees the development of programs and resources supporting institutionally related foundation boards, institutional governing boards, and other senior staff and volunteer leaders involved in higher education fundraising and stewardship. David previously served as AGB’s director of foundation programs and research.

Mark Hobbs
CEO
Fundmetric
Mark Hobbs is known for his innovative approach to leveraging technology in non-tech industries and has played a pivotal role in Fundmetric’s journey, guiding the organization from its inception. He has co-written several peer-reviewed publications including a paper for the International Electrical Engineers (IEEE) on Machine Learning the Donor Journey. His work is featured in multiple textbooks on Machine Learning. He works closely with clients to create award-winning campaigns and is focused on the lifetime value of every donor.

Fundmetric

With Thanks to AGB Mission Champion:
Fundmetric

Subscribe to AGB podcasts on these platforms:

Apple Podcasts   Spotify   Stitcher   Audible

The owner of this website has made a commitment to accessibility and inclusion, please report any problems that you encounter using the contact form on this website. This site uses the WP ADA Compliance Check plugin to enhance accessibility.