TAAS 2020 Speakers
The theme for TAAS 2020 is “Advancement Analytics – The Current State and Future Opportunities“. Speakers and topics have been selected that address this theme to create a program that focuses on both theory and practice. We look forward to the conversation!
2020 Vision: Observations on how Artificial Intelligence is Disrupting the Business of Philanthropy
Advances in artificial intelligence and machine learning are accelerating the business of
philanthropy, and are especially relevant to the field of higher education which has a deep history of leveraging information to increase outcomes. This paper aims to describe the current and future state of advancement analytics, and explores a variety of questions on the horizon. What can a 1991 case study about Lotus Notes teach us about analytics implementation in 2020? What would change if analytics were used more to support the work of front-line fundraisers, rather than punish them? What if advancements in machine learning were applied to all levels of the giving pyramid, not just to identify the few constituents at the top? What is the impact on large, decentralized institutions, who need to reconsider their own internal policies on how prospective donors are discovered, qualified, and asked? Additional considerations include how our industry might learn from the successes and missteps observed in the private sector.
Melissa Cox is currently the Lead Director of Annual Giving for the Ross School of Business at the University of Michigan (U-M). In 13 years at U-M, Melissa has served in a variety of fundraising and operational roles and was a member of the Campaign, Information, and Technology leadership team during the Victors for Michigan campaign which ultimately raised over $5 billion from nearly 400,000 donors.
Melissa earned a Master of Science in Information from the University of Michigan’s School of Information and a Bachelor of Science in the theory and practice of social change from the University of California Santa Cruz. They are a certified SCRUM Master, and their work has been recognized by the Council for Advancement and Support of Education (CASE) Circle of Excellence with two Gold Awards. Additionally, Melissa volunteers as an advocate and mentor for U-M’s Development Summer Internship Program, which prepares undergraduate interns for careers in philanthropy.
Melissa credits a temporary job at a toilet factory as their introduction to the wonderful world of relational databases, kickstarting a passion for finding solutions at the intersection of people and technology. When they’re not overthinking how artificial intelligence is disrupting the business of philanthropy, Melissa enjoys reading academic papers about the Terminator franchise.
Leveraging Insights from Other Industries to Accelerate Data-Driven Success in University Fundraising
It is our expectation that the changes that have occurred in other industries will be mirrored over the next several years in higher education. University Fundraising Organizations who recognize these similarities will be uniquely positioned for great successes in a rapidly changing business environment. This paper provides unique insights into parallel functions in other industries by examining four key trends:
- The Evolving Role of Analytics Leadership
- The Explosion of New Data Related Solutions
- Headcount Changes: Separating Fact from Fiction
- The Immediate Need for a Robust Data Strategy
Jim Dries is the CEO of piLYTIX, an Austin based artificial intelligence tech company that generates revenue-enhancing insights for users in several industries. Jim credits piLYTIX’s success to the company’s team of exceptional data scientists and a collaborative client base that strives to be difference makers for their organizations.
Jim’s 20-year career has spanned leadership roles in finance, product development, sales and marketing. Until recently, however, none of those roles intersected with the world of university fundraising. As piLYTIX began working with collegiate sports teams to improve the efficiency of their ticket sales organizations, several university clients inquired whether piLYTIX’s A.I. engine could be leveraged to drive similar efficiencies in their fundraising organizations. In 2019, The University of Texas – Austin became the first university to implement piLYTIX’s A.I. software to complement the work of its internal fundraising analytics team.
Jim is now on a mission to elevate the role of the analytics team in higher education fundraising. As fundraising organizations are called on to achieve greater financial targets than ever before with tightening budgets, Jim believes that smart, capable analytics leaders armed with cutting edge toolsets need to be seen as the driving force that guide every organization’s fundraising strategy.
Originally from Wisconsin, Jim has degrees from Yale University and the University of Chicago. Jim and his family live in Austin, TX, where piLYTIX is headquartered.
Amanda Jeppson & Rebecca Stapley
Hidden Voices: Using Social Listening to Uncover Audience Insights and Boost Advancement
As the higher education landscape changes and new challenges evolve, especially difficulties related to funding, this paper explores how colleges and universities can harness the innovative power of social listening to better understand target audiences, to build and maintain affinity and sense of community, and to ultimately support advancement efforts. Social listening is the research approach of processing online mentions of a specified topic and distilling that data into actionable insights to guide data-driven decisions. Using Midwest College (a pseudonym) as a case study, this paper illustrates the types of data observations and subsequent strategic insights gleaned from social listening that can be applied to benefit advancement efforts. Specifically, these insights are recommended in the context of affecting campaign communications, marketing, and potentially gifting behaviors. While the specific findings herein may not be generalizable to college campuses outside of Midwest College, social listening as an overarching approach to advancement in higher education is universal in its potential to unlock complex findings and unearth context-specific revelations that power advancement offices and their approach to sustained affinity, sense of community, and shared experience. With alumni giving as the second largest source of funding to colleges and universities, colleges and universities cannot afford to miss the opportunity to simply listen.
Amanda Jeppson is a Social Media Data Analyst at Campus Sonar, a social listening agency primarily serving higher education institutions, where she collects data from the far-flung reaches of the internet, analyzes it, and reports the story the data tells her. She, along with the other Campus Sonar Strategists, delivers powerful data-driven insights to inform college and university administration decision-making.
With more than seven years of research experience, largely focused on qualitative or mixed methods research approaches, she knows and embraces the power of data to catalyze meaningful change. Having witnessed the birth of Facebook, survived competitions for her friends’ Top 8 spots on MySpace, and mastered the gif game and meme madness, she also understands the importance and pervasiveness of social media and online life to communicate both information and personal identity. What happens online is real life, and Amanda now spends her time harnessing online data for the better. Published in the GreenBook blog and presenting a poster at the American Marketing Association Advanced Research Techniques Forum, Amanda eagerly strives to learn and share the benefits of leveraging social and online data by itself and in conjunction with traditional research methods. A nerd by day and by night, Amanda enjoys exploring the landscape of data sources to better understand the world and how it works.
Prior to Campus Sonar, Amanda pursued her passion of higher education research at the Wisconsin Center for Education Research in Madison, Wisconsin, exploring topics such as how professors teach, student study habits, and the manufacturing and biotechnology skills gap in Wisconsin. Her work ultimately resulted in the publication of Beyond the Skills Gap and a handful of conference presentations and peer-reviewed publications. To better understand the earlier end of the education pipeline, she worked as a Qualitative Data Analyst in the Research and Program Evaluation Office within the Madison Metropolitan School District, where she managed qualitative data analysis for key projects such as the Strategic Framework and the yearly Climate Survey, while also completing smaller research projects in collaboration with schools and other Central Office departments.
Amanda holds a Master of Arts degree in Educational Policy Studies from the University of Wisconsin–Madison, where she also earned her Bachelor of Arts degree in English and Psychology.
Rebecca is a Strategist at Campus Sonar. She has spent more than nine years in higher education enrollment and marketing, honing her skills and passion for delivering results driven strategy that aligns with institutional goals and values. While on campus, Rebecca piloted a student blog targeted towards prospective students and managed a team of student content creators who added their authentic voice to help tell the college story. She also served as strategist and project manager for the college’s Giving Day and launched a new strategic partnership between alumni and marketing focused on social media. Rebecca holds a Master’s Degree in Integrative Marketing and Communications and is a noted public speaker who has presented at various higher education conferences.
Data science for Fundraising: a Review of Analytics in Fundraising
Data analysis, data mining, predictive analytics, machine learning, data science, and
artificial intelligence have affected how for-profit organizations make decisions for many years
now, many decades for some methods. The nonprofit industry, specifically, nonprofit
fundraising has been trying to catch up. This paper lists various applications of analytics in
nonprofit fundraising as found in the literature. I present the survey in two ways: a)
chronological, by decades and b) by analytical methods. I use the term “analytics” to
capture the various methods used in data mining, predictive analytics, machine learning,
data science, and artificial intelligence. This paper is structured as follows: 1) a brief review
of the analytics methods, 2) review of the literature by decades, 3) review of the literature by
methods, 4) summary of the findings, and 5) predictions on future work.
A co-author of Data Science for Fundraising, an award-winning keynote speaker, Ashutosh R. Nandeshwar is one of the few analytics professionals in the higher education industry who has developed analytical solutions for all stages of the student life cycle: from recruitment to giving. He enjoys speaking about the power of data, as well as ranting about data professionals who chase after “interesting” things. He earned his Ph.D./MS from West Virginia University and his BEng from Nagpur University, all in industrial engineering with an emphasis on artificial intelligence. Currently, he is leading the data science, reporting, and prospect development efforts at the University of Southern California.
When XGBoost Tree Meets Humanity: An Exploration of Implementing New Machine Learning Techniques into Every Day Fundraising Practice
As the discipline of data science grows, more and more complex algorithms get applied to an ever larger variety of – and size of – data sets. The good news is that an R Square value of .97 is now plausible without the help of omniscience. The bad news is that a machine learning algorithm can now produce 500 rules from a Random Forest model, and those 500 rules need to be programmed (often by hand) into an organization’s data mart. This paper explores options for either adding complex algorithms to a CRM or for using tributary programs to score segments and deliver them back into an organization’s reporting system.
Marianne has worked in the nonprofit industry for over 30 years, principally in prospect research and prospecting. Her career includes establishing prospecting programs, introducing analytics to her employers, running an annual giving program with an average increase of 27% per year, and teaching others how to use machine learning and donor modeling to improve results. She co-founded Staupell in 2012 and now offers data management, optimization, and analysis to clients, as well as training. Her workbook, Building Your Analytics Shop: A Workbook for Nonprofits, is available on staupell.com and on Amazon.
Predicting Capacity With Machine Learning
Capacity ratings in prospect research have traditionally relied on aggregated wealth surveys to infer a prospect’s total wealth based on the prospect’s publicly available data. Such surveys typically report the average asset breakdown by total net worth, which makes it impossible to reclaim the individual-level wealth data. Using the row-level wealth data reported by the Survey of Consumer Finances (SCF), I demonstrate a supervised machine learning approach to capacity estimation that utilizes the XGBoost algorithm. This approach acknowledges the uncertainty inherent in a capacity prediction in providing a range of ratings and their associated probabilities. I will show that this approach results in a higher accuracy rate than a capacity formula based solely on real estate holdings.
David Schemitsch is a data science engineer on the Insights & Analytics team at Columbia University’s Office of Alumni and Development. He joined Columbia’s Prospect Development team as a research analyst in 2013. He began his career in nonprofit fundraising in 2009 at the International Rescue Committee, where he served in prospect research and advancement roles. He received a BA in politics and anthropology from NYU and is currently pursuing an MS in data science from Columbia University. He has previously presented on using and learning R at Apra events and conferences.