AI + Fundraising Webinar

Below are examples of AI tools, platforms, and vendors that can help with specific fundraising initiatives and processes. When selecting AI tools, consider factors such as integration capabilities with your existing systems, scalability, user-friendliness, and cost-effectiveness to ensure they align with your organization’s specific needs and objectives.

• Meeting transcription and CRM integration tools include Otter.ai,  Fireflies.ai, and Avoma.
• Predictive modeling for identifying new prospects can be achieved through programs like Python and R. To learn more about this work and how it can help you achieve your fundraising goals, check out our case studies or contact us.
• Content generation for text and image-based communications can be aided with tools such as ChatGPT/OpenAI, Claude, Google Gemini, and Microsoft Co-Pilot.

You can start now! Generative AI tools can help you be more efficient in creating solicitation emails and stewardship and cultivation letters. Even for prioritizing donors, you can start with simple engagement scores that can be generated in a spreadsheet. That being said, AI is most effective when it can work with well-structured, relevant data, so if your organization lacks sufficient data, focus on improving data collection processes first (e.g., capturing donor interactions, tracking event attendance, or logging campaign outcomes). Define specific areas where AI can make a significant impact (e.g., predicting donor churn, personalizing outreach, or automating reporting), and calculate potential ROI based on time savings, cost reductions, or improved fundraising outcomes. You can educate the rest of your organization on these points, and ensure that this type of collaboration extends across departments, particularly IT and Legal as it concerns data security. In adoption, nonprofits should start small, with a pilot experiment to ensure controlled use, evaluate scalability, and create guidelines or policies for safe use.

This is an important issue that many organizations are dealing with. One idea is to use AI tools embedded in internal systems to analyze data without exposing it externally, and/or partner with vendors offering robust security and compliance measures. You want to avoid sharing sensitive data with external Generative AI tools. Your organization should define clear policies on what AI can and cannot be used for, approved tools, and compliance measures. It should also conduct training for all staff and maintain transparency in how AI handles data.

We believe you should focus on high-impact and low-cost or free tools, like ChatGPT, Grammarly, or Canva, to enhance donor communications, draft outreach emails, and create visuals. You can also leverage AI features in your existing CRM, such as Salesforce Einstein or HubSpot’s AI tools, for donor segmentation and engagement insights. If your data is well-structured, explore simple AI-driven donor segmentation or predictive modeling to prioritize high-potential prospects. As with any AI adoption, start small, scale gradually, and measure your outcomes to justify further investment.

There are many applications of AI to help improve the ROI of direct mail initiatives, including your annual fund. Predictive AI models can be built that predict the likelihood of someone making a gift for an upcoming mailing. They can also be used to estimate the optimal ask amounts for each prospective donor. These types of giving inclination scores and ask amount values can be combined to help prioritize mailing lists so that the mailing generates an optimal ROI. In addition, Generative AI tools can be used to help create draft letters and refine communication. Furthermore, Unsupervised Learning tools, such as K-means Clustering, can be used to segment populations and identify the key characteristics of those populations.

This is an important question. The key to successfully using Generative AI is to create and submit effective prompts that are specific, including context, purpose, audience, and desired outcome. Ask a clear question and set boundaries, such as “provide 10 bullet points outlining XYZ’s impact on ABC group.” If there are keywords you want to include, make sure you include them in your question. Finally, feel free to refine the output, for example “expand this explanation with 3 examples” or “say this in more casual language.”

There are not a lot of off-the-shelf AI tools that conduct relationship mapping, as data is required to explain how many degrees of separation are between two people. Tools that provide this type of service include RelSci, LexisNexis, WealthEngine, DonorSearch, and iwave.

You should have this in your AI policy, so everyone knows how to best communicate this information, as well as who is a resource for additional questions. Here is an example: “Our organization is committed to transparency and responsible use of technology in our fundraising efforts. We use artificial intelligence (AI) tools to enhance the efficiency and personalization of donor communications, such as drafting emails, generating campaign materials, and analyzing donor engagement trends. However, these tools are employed solely as aids to improve our outreach and operational efficiency; all final messages are reviewed and approved by our team to ensure alignment with our mission, values, and the trust donors place in us. We take donor privacy and data security seriously. AI tools are used in compliance with strict data protection policies, ensuring that no sensitive or personally identifiable donor information is shared with external systems. Should donors have any concerns or questions about our use of AI, we encourage open communication and will provide detailed information about the steps we take to ensure ethical and responsible practices in all aspects of our work.”

RAG – retrieval augmented generation – would be an effective solution for this. This is an approach where information retrieval is combined with generative models to improve the quality and relevance of generated outputs. It involves:

1. Retrieval: The system searches and retrieves relevant information or documents from a database or knowledge base to provide accurate and context-specific data.
2. Augmentation: The retrieved data is used to enhance the generative process by giving the AI model additional context or facts.
3. Generation: A generative AI model, such as a large language model, uses the augmented information to create more informed and precise responses.

This approach is particularly useful in tasks requiring accurate and up-to-date knowledge, as it combines the strengths of retrieval systems (like search engines) with the creative and flexible capabilities of generative AI models.