AI is Changing Fundraising. Just Not in the Ways Most Organisations Think.

By Tobin Aldrich

Following the launch of AAW’s AI CAM, it has certainly got conversation flowing. Ask most fundraising directors what they are doing with AI right now and you will hear a familiar set of answers. Better email targeting. Smarter segmentation. AI-assisted content. Prospect screening tools. All of it is real and actively being invested in. Much of it is delivering modest gains, but very little of it is strategic.

The sector is currently in a wave of tactical AI adoption. Individual teams responding to individual challenges and buying individual tools. The direct marketing team uses AI to improve open rates. The major gifts team has invested in screening platforms. The digital team is experimenting with AI-generated content. Each decision makes sense in isolation. Taken together, they add up to high cost, added complexity and a series of improvements that sit comfortably within existing fundraising models rather than challenging them.

This is the central argument of this series. AI is a genuinely transformative technology, arguably the most significant shift the sector has seen since the advent of the internet. But the transformation it enables is not about doing the same things more efficiently. It is about doing things differently. Right now, most organisations are investing in the former while missing the latter.

The Targeting Trap

Take email as an example. AI targeting tools have become one of the most widely adopted fundraising technology investments in recent years. The promise is clear: use machine learning to identify the right supporters, at the right time, with the right message. In practice, these tools often deliver exactly that. Targeting improves. Waste is reduced. Cost per response falls.

But there is a harder question that is rarely being asked. If open rates are sitting at around 18% and falling, unsubscribe rates are rising, and donors consistently report feeling over-contacted and under-valued, what problem is better targeting actually solving?

In many cases, it becomes an exercise in refining the delivery of something audiences are increasingly disengaged from. The issue is not who receives the message. It is whether the message itself still resonates.

This is not a critique of targeting technology. It is a critique of using it as a substitute for a more fundamental issue. Many charities do not have a data problem. They have a donor relationship problem. Supporters who feel genuinely connected to a mission, who understand impact and feel known rather than processed, do not need better targeting. They simply want to hear from you.

What AI is Actually Doing Well Right Now

There are areas where AI is already delivering meaningful value in fundraising, and they are worth recognising.

In prospect research and major gift fundraising, AI tools are changing the economics of identifying and prioritising potential donors. Work that once took days of manual research can now be completed in minutes, often with broader coverage and improved accuracy. Organisations using these tools well are finding prospects they would previously have missed.

Predictive modelling for retention is also having an impact. AI can identify donors at risk of lapsing with a level of precision that manual analysis struggles to match, allowing organisations to intervene earlier. Given the cost difference between retention and acquisition, this is significant. However, the tool alone does not address why donors lapse in the first place.

AI is also changing how people discover charities. Increasingly, supporters are using tools like ChatGPT and Claude to research causes rather than relying on search engines. This is already reducing direct traffic to charity websites, and the trend is likely to accelerate. Organisations that cannot clearly express their mission in ways these systems can interpret risk becoming less visible to potential supporters, not because they are doing anything wrong, but because the environment is shifting.

The Piecemeal Problem

The common thread across most AI activity in the sector is that it is emerging from the bottom up rather than the top down. Individual teams identify individual problems and adopt individual solutions. There is rarely anyone stepping back to ask a more fundamental question: what are we actually trying to achieve, and are these tools creating a coherent whole or just a collection of disconnected investments?

The result is predictable. Duplication of spend. Systems that do not integrate. Data trapped in silos. Strategic decisions about AI are being made implicitly rather than deliberately.

This is where leadership becomes critical. The question for CEOs and Fundraising Directors is not what AI tools they should be adopting. It is what kind of fundraising organisation they want to become over the next five years, and what role AI should play in that direction. These are very different questions, and the second is still not being asked often enough.

The next article in this series looks at why this tactical approach is not just limited but increasingly risky, and what needs to be in place for a more strategic model to emerge.

This is the first in a series of three articles on AI and the future of fundraising.