A donor gives $100.

She cares about your mission. Something moved her enough to act. She gets a receipt, maybe a thank-you email, and then life keeps moving.

So does your organization.

There’s an event coming up. Year-end planning starts. Someone leaves the team. A grant is due. The newsletter needs to go out. Your development staff has a list of people to call that never quite gets finished.

The donor still cares.

But a month passes. Then three. Then six. Eventually, your database gives her a new name:

Lapsed.

She didn’t necessarily decide to leave. The relationship just got quiet.

That tension is why I’m interested in fundraising automation. Not because I think fundraising needs fewer people. I think the opposite.

For years, technology has helped fundraisers send more emails, process more gifts, organize more data, and run more campaigns. AI allows us to think bigger.

What if the next era of fundraising technology isn’t about automating more communication? What if it’s about helping our teams notice more people?

In this post, I’m going to cover:

• Why donor retention is increasingly a capacity problem

• Why campaigns may be the wrong organizing unit for engagement

• What I mean by an AI engagement team

• A four-question framework for identifying the moments that matter

• Where data, AI, and people each belong

• What we’re learning from real engagement campaigns

• Why the future of fundraising automation should be more human, not less

Fundraising doesn’t have a care problem

I’ve spent a lot of my career around nonprofit and ministry teams.

I’ve never heard a fundraiser say, “I wish we cared less about our donors.”

Most teams know exactly what they would like to do.

Call more donors. Thank people faster. Follow up after the first gift. Share impact. Ask questions. Reach people before they lapse.

The problem is doing it consistently.

And the numbers suggest that problem matters.

The Fundraising Effectiveness Project reported that total charitable dollars increased an estimated 5% in 2025 while the number of donors declined 3.6%. In other words, organizations raised more money from a smaller donor base.

The relationship is especially fragile early on. M+R Benchmarks 2026 found that just 24% of its defined cohort of new online donors returned to make another online gift the following year, compared with 66% of prior donors.

Yet this isn’t primarily a knowledge problem.

In Bloomerang’s Mission Retainable research, 82% of fundraisers said limited staff and resources were their biggest obstacle to improving retention. Sixty-five percent of donors wanted regular updates about their impact, while only 36% of nonprofits said they provided them.

The desire is there on both sides.

What’s missing is capacity. In other words, more relationships than hours.

Campaigns are partly a product of that constraint

Fundraising has traditionally been organized around campaigns.

Spring appeal. Giving Day. Gala. Renewal campaign. Year-end.

That makes sense. Human teams need beginnings and endings. We need a list, a message, a deadline, and a manageable workload.

But donors don’t experience their relationship with an organization as a campaign calendar.

They experience moments.

They make a first gift.

They attend an event.

They answer a survey.

They miss a recurring payment.

Their giving pattern changes.

They ask someone to call.

They quietly approach one year since their last gift.

The campaign has been a useful unit of work for the organization. The moment is the donor’s more natural unit of relationship.

AI changes how we organize around those moments.

That may be the most important opportunity in fundraising automation.

Instead of asking only:

What campaign are we sending next?

We can increasingly ask:

Who needs attention next?

That’s a different operating model.

We don’t have a message problem. We have a moment problem.

Fundraisers spend an enormous amount of time thinking about messages.

What’s the subject line?

What’s the appeal copy?

What story should we tell?

Those things matter. But a perfect message sent at the wrong time to the wrong person is still the wrong message.

Before asking what to say, I think teams should answer four questions.

1. Who needs attention?

Look for a person or group where something meaningful has happened.

A first-time donor. An event attendee. A longtime donor whose giving changed. Someone approaching lapse. A supporter who raised a hand and requested more information.

2. Why now?

Outreach should have a reason.

A gift happened. An anniversary is approaching. A payment failed. Someone responded. An event is coming up. A behavior changed.

Timing gives the engagement context.

3. What should happen next?

Not every signal requires another solicitation.

Sometimes the next step is a thank-you. An impact update. A question. A reminder. A survey. A useful link. A phone conversation.

The goal is movement in the relationship, not simply another message.

4. When should a person take over?

This is the question every automation needs.

If the conversation becomes sensitive, complicated, high-value, unusual, or deserves personal attention, route it to a person.

A useful engagement system should not create distance between donors and staff.

It should surface the relationships that deserve more human attention.

For a tactical version of this framework, we’ve written separately about building a donor engagement system around what should happen next.

The new model: data, AI, people

I think the emerging operating model can be summarized in three parts.

1. Data tells us who

A donor database is great at telling us what happened.

Someone gave $100 on March 14. Someone attended an event. Someone hasn’t given in 10 months. Someone has supported the organization for five years.

The next step is using those signals to determine who may need attention.

You don’t necessarily need sophisticated predictive analytics to begin. Start with the information already in your system.

Who gave for the first time this month?

Who normally gives every six months but hasn’t?

Who attended an event and never received personal follow-up?

Who is 10 or 11 months removed from a previous gift?

The data helps answer who and why now.

2. AI gives us capacity

Once you know who deserves attention, someone still has to do something.

That’s where an AI engagement assistant can help.

Appropriate, structured outreach can happen through voice, text, email, surveys, reminders, and follow-up workflows without requiring staff to execute every individual touchpoint manually.

That doesn’t mean handing the relationship over to AI.

It means using technology for the repeatable work around the relationship so the team has more capacity for the work that requires people.

We’ve written more about this distinction in our guide to practical AI voice use cases for nonprofits.

3. People build relationships

This is the part I don’t want us to automate away.

A donor asks a complicated question.

Someone wants to increase their support.

An alum wants to reconnect with the university.

A longtime supporter says they’re going through a difficult season.

Someone says, “I’d like to talk to somebody.”

A person takes over.

Data tells us who. AI gives us capacity. People build relationships.

I think of that as an AI engagement team. EverRaise is the AI engagement assistant that helps the human team put that model into practice.

We’re already seeing the operating model change

You don’t have to look far to see organizations rethinking traditional engagement programs.

Consider university phonathons.

For decades, the model was fairly simple: put students in a room, give them lists, and work through the phones.

Mercer University now operates what it calls a Digital Engagement Center. Launched in 2024, the center restructured Mercer’s previous student calling program around what the university describes as a combination of technology, data, and human touch. Student engagement officers still make calls, but they also create personalized videos and emails for alumni and friends, with a stated focus on building relationships.

The interesting part isn’t simply that Mercer added email and video. It’s that the university restructured the traditional calling program around a broader engagement model.

They changed the operating model around the conversation.

We are seeing something similar in our own work at EverRaise.

In one recent university engagement campaign, EverRaise reached 868 people in less than five hours, generating 163 live answers and 116 link clicks while saving more than 20 hours of estimated staff time. We documented those results in our September EverRaise product update.

The most interesting number to me isn’t 868.

It’s the 20-plus hours.

Because capacity changes what a team can do next.

Instead of spending those hours dialing, logging unanswered calls, sending individual follow-ups, and working down a list, staff can spend more r time on the people who answered, responded, showed interest, or actually need a conversation.

That’s the point.

The efficiency is not the relationship. The efficiency creates room for the relationship.

Start with moments, not an AI strategy

You don’t need to go back to your organization and create an AI transformation committee.

Start smaller.

Pick one meaningful moment where your team regularly loses the thread.

Maybe it’s a first gift.

Instead of saying, “We need to do a better job with first-time donors,” define the workflow.

A gift comes in.

What happens in 24 hours?

What happens after a week?

What happens after 30 days?

What information should change the path?

What response requires a staff member?

Then try another moment.

Someone is 10 months removed from their last gift.

Don’t wait for the database to call them lapsed.

Thank them. Check in. Remind them what their support made possible. Give them an appropriate opportunity to reconnect.

As I wrote in our guide to donor retention strategies, consistent engagement turns retention from a report into actual work.

This is where fundraising automation becomes much more useful.

Not:

Automate everything you can.

Instead:

Identify the moments that matter, automate the work that can responsibly scale, and protect the places where people matter most.

The future isn’t automated fundraising

That’s not the future I want to build.

AI used poorly can give organizations the ability to create more noise faster. More generic emails. More unnecessary calls. More communication nobody asked for.

We don’t need that.

The opportunity is to build systems that make organizations more attentive.

Imagine that donor again from the beginning.

Ten months after her gift, her name surfaces as she approaches the anniversary of her last contribution.

She gets a thoughtful call.

Maybe she answers. Maybe she doesn’t. If she doesn’t, she receives an appropriate follow-up message.

She responds:

“I’ve actually been meaning to reconnect. Could someone call me?”

And someone does.

She didn’t need your organization to have a sophisticated AI strategy.

She needed someone to remember her.

That’s the future of fundraising automation I find compelling.

Not fewer fundraisers.

Not automated relationships.

More capacity to notice.

More capacity to listen.

More capacity to follow through.

More human attention where human attention matters.

Data tells us who. AI gives us capacity. People build relationships.

And if we build this well, AI’s greatest contribution to fundraising may not be helping us communicate with more people.

It may be helping us remember more of them.