AI-assisted donor engagement: an illustrative scenario

The situation

A not-for-profit holds donor information in a customer relationship management (CRM) system, an email tool and event spreadsheets. A small team sends mostly similar appeals to broad lists because segmenting by hand takes too long, and it is hard to tell which supporters may be drifting away.

Trust is the organization's most important asset, so any use of donor data must feel appropriate to the people who gave it.

What the organization wants to achieve

• Understand supporter journeys and likely next steps.
• Tailor outreach so it is more relevant without feeling intrusive.
• Free staff time from manual list building.
• Respect consent and communication preferences throughout.

Approach

ARCHR would typically run this through its four phases: Discovery and Strategy, Pilot and Validation, Build and Scale, then Optimize and Evolve. In this scenario that could look like:

• Discovery: review data sources, consent records, current campaigns and the questions fundraisers most want answered.
• Pilot: clean and join a subset of donor data, then test simple segments and propensity indicators on one campaign.
• Pilot: compare tailored outreach against the regular approach in a fair test.
• Build: resolve duplicates into one master record per supporter, keep consent and communication preferences with that record, and automate data refresh into a small, low-cost warehouse with a semantic layer for self-service reporting.
• Build: document every segment, add consent checks before any message is generated, and keep a person in the loop to approve generated copy.
• Optimize: review results with fundraisers, retire what does not help and share learnings with the board.

Technology and governance considerations

A modest data platform on AWS, Microsoft Azure or Google Cloud, whichever the organization already uses, could keep costs in proportion to a nonprofit budget, with FinOps reviews to watch spending. Donor data would be minimized, protected by role-based access and retained only as long as needed, consistent with the privacy law that applies, such as Canada's Personal Information Protection and Electronic Documents Act (PIPEDA), the European Union's General Data Protection Regulation (GDPR) or state laws.

Resilience and ownership matter even for a small team. Backups, documented pipelines and a named data steward would keep the setup maintainable if staff change, and the board would receive a short summary of how donor data is used and protected. Where volunteers or partner organizations touch the data, their access would be limited and reviewed regularly.

Models would be explainable enough for fundraisers to challenge, and staff would review messaging before it goes out. Sensitive attributes would be excluded unless there is a clear and appropriate reason. Generated copy would follow the organization's voice and be approved by a person.

What success could look like

A pilot would agree measures and a baseline before any work starts. Qualitative signals the team could track include:

• Fundraisers can explain why a supporter received a given message.
• Staff spend less time on manual list preparation.
• Supporters' preferences and opt-outs are respected and visible.
• Test results give the team honest evidence about what works.

Actual results depend on an organization's data, systems and objectives. This page is not legal, financial or investment advice.