Your Next Best Action Might Be No Action at All

What our obsession with moving relationships forward may be getting wrong

In fundraising and engagement, we spend a lot of time thinking about what should happen next. Someone makes their first gift, attends an event, signs a petition, watches a video, responds to a survey, visits a planned giving page, or suddenly becomes more engaged with our content, and each new signal prompts a familiar question: What should we do now?

I understand the instinct. In fact, this post was inspired by a recent conversation that made me recognize how deeply ingrained it is in my own work. I work in sales and marketing, where “what’s the next step?” is practically a reflex. We’re trained to look for movement, create momentum, and figure out how to advance a relationship toward some desired outcome.

Nonprofits operate under many of the same pressures. Fundraising goals need to be met, campaigns need to perform, donors need to be retained, and relationships need to be cultivated. So we build strategies and systems designed to move people toward something, then create metrics to tell us whether that movement is happening. All of that makes sense. But the conversation got me thinking about whether our desire to see progress can sometimes become more about satisfying our own process than responding to what the relationship actually needs.

Lately, though, in my recent posts about aligning your organization around the donor experience, I’ve been wondering whether our constant desire to move relationships forward sometimes causes us to move them before they’re ready. There are moments when someone has clearly demonstrated interest or affinity, but the situation doesn’t necessarily call for another communication, another ask, or another carefully orchestrated next step. Sometimes the relationship may be exactly where it needs to be for the moment. Which raises a slightly uncomfortable possibility: sometimes the best next action may be no action at all.

We’ve Turned Relationships Into Journeys

Nonprofits have become increasingly sophisticated about designing constituent journeys, and for good reason. A first-time donor shouldn’t receive the same experience as a long-time sustainer. A volunteer who becomes a donor brings a different history to the relationship than someone who arrived through an acquisition campaign. Someone who attends events, consumes content, advocates for the mission, and occasionally gives shouldn’t be understood solely through their most recent transaction.

Thinking about those experiences as journeys has helped organizations become more intentional about the relationship rather than simply reacting to individual transactions. But journeys also come with an implicit expectation of movement. A first gift should lead to a second. A second should lead to recurring giving. A recurring donor might become a mid-level donor, who might eventually become a major-gift prospect. An advocate becomes a donor. An event participant becomes something else.

We’ve taken something fundamentally messy and nonlinear—a human relationship—and created stages around it so we can manage, measure, and improve it. Organizations need to do that. The problem begins when the journey we’ve designed starts to matter more than the person actually taking it.

That’s when a subtle shift happens. Instead of asking, “What does this relationship need right now?” we begin asking, “How do we get this person to the next step?” Those questions can produce the same answer, but they begin from very different places. One starts with the person and the relationship. The other starts with the outcome we’ve already decided we want.

Not Every Signal Requires a Response

Consider someone who has supported an organization for years. They give consistently, read the newsletter, occasionally attend an event, and clearly care about the mission. One afternoon, they spend time reading several articles about a particular program. That’s useful information. It tells us something new about their interests and adds another dimension to our understanding of the relationship.

There may be a genuinely relevant reason to follow up, but our increasingly sophisticated engagement infrastructure tends to assume that some kind of response is preferable to none. The website visit becomes a signal, the signal meets a threshold, the threshold triggers a journey, and suddenly the constituent is receiving something because the system has been designed to make something happen. Meanwhile, they may have simply wanted to learn more about a program they care about.

I think we sometimes make the same mistake more broadly by treating attention as an invitation for more attention. Someone reads something, so we recommend something else. They attend an event, so we begin another cultivation sequence. They click, so we retarget. Then they give, and before the thank-you has even landed, we’re thinking about the next gift.

Those interactions absolutely matter, and each one can help us understand the relationship more completely. But perhaps not every interaction needs to be converted into momentum. Sometimes the most audience-first thing we can do with a signal is simply remember it, allow it to become part of what we know about that person, and use it when the relationship actually calls for something from us.

Next Best Automation Has a Bias Toward Action

This is where technology creates an interesting challenge, because most automation is built around a simple construct: if something happens, do something. A donor makes a gift, so a stewardship journey begins. Someone attends an event, so a follow-up sequence starts. A constituent crosses an engagement threshold, so they enter a new audience. A donor hasn’t given recently, so a reactivation campaign kicks in.

We spend enormous amounts of time designing these workflows, identifying triggers, defining audiences, and determining what happens next. I’m not sure I’ve ever seen someone proudly demonstrate a sophisticated automation workflow where the intended outcome was Nothing Happens. Maybe we should.

Imagine an engagement model that recognizes a new signal and considers it alongside everything else we know about the constituent. The system has done exactly what we want it to do: it has helped us better understand the person and the relationship. But instead of automatically translating that knowledge into another communication, it determines that nothing we could do right now would meaningfully improve the relationship. No email is triggered, no task is created, and no solicitation follows. The signal remains valuable because we learned something, even though nothing happened as a result.

That’s not neglect. The constituent wasn’t forgotten, and the opportunity wasn’t missed. It’s intentional restraint—the recognition that knowing something new about a relationship doesn’t automatically create an obligation to act on it.

AI Changes the Equation

For years, nonprofits had a natural constraint on how much they could act on constituent behavior: capacity. Creating communications took time, developing segments required work, personalizing content required effort, analyzing behavior required analysts, and building and managing journeys required people. Organizations couldn’t respond differently to every signal because doing so simply wasn’t practical.

AI is rapidly changing that. We’re gaining the ability to analyze far more signals, identify smaller audiences, generate more content, personalize more interactions, predict more behaviors, and automate more decisions at a scale that would have been impossible only a few years ago. There is enormous potential in that, particularly when it helps nonprofits understand people more completely and create experiences that are genuinely more relevant.

But removing the friction from communication also removes one of the things that used to force us to exercise restraint. We’re approaching a world in which it may be technically possible to respond to nearly every constituent signal, and that makes the question of whether we should respond much more important. Otherwise, we risk becoming exceptionally good at producing highly personalized, perfectly timed, completely unnecessary communications.

That’s why I think the idea of “next best action” needs to evolve along with the technology. A sophisticated model shouldn’t simply choose whether the next interaction should be an email, text, direct-mail piece, phone call, solicitation, content recommendation, or event invitation. If the model is genuinely considering the health and context of the relationship, it should also be capable of determining that none of those things would improve it right now. “Not yet” should be a legitimate next best action.

Maybe Restraint Is a Form of Personalization?

We usually describe personalization in terms of what an organization knows about someone: their giving history, interests, engagement patterns, program affinities, communication preferences, and increasingly, the many ways they interact with the organization beyond fundraising. The assumption is that knowing more allows us to communicate more effectively, and I think that’s true.

But perhaps knowing more should occasionally allow us to communicate less. If we genuinely understand the relationship, we should be better able to distinguish between a meaningful opportunity to engage and a moment when someone is simply interacting with an organization they care about. We should know when a stewardship touch would feel thoughtful, when an invitation would be relevant, when an ask makes sense, and when another communication would mostly serve our own desire to create movement.

That strikes me as a more meaningful definition of personalization than simply getting better at deciding which message to send. The organizations that know their constituents best shouldn’t necessarily be the ones that communicate with them most frequently. They should be the ones exercising the best judgment about when they have something relevant and worthwhile to say.

There’s also something deeply audience-first about accepting that a healthy relationship can include periods when nothing much happens. A donor can remain deeply committed to a mission without responding to the latest campaign. A volunteer can care enormously without becoming a donor. Someone can attend an event without needing to be cultivated into something else. People can pay attention, learn, participate, give, step away for a while, and return later. Those pauses are part of relationships, too.

What Does the Relationship Need Right Now?

I still believe strongly in intentional donor journeys, cultivation, stewardship, thoughtful next steps, and using data to understand how relationships are developing. This isn’t an argument for sitting around and hoping donors eventually raise their hands, nor is it an argument against using the increasingly sophisticated technology available to us.

It’s an argument for remembering what all of that technology is supposed to serve. A constituent relationship doesn’t exist to satisfy the stages we’ve designed for it. Our journeys, segments, engagement scores, predictive models, and automation should help us understand and support relationships, not force those relationships to move according to our timetable.

So perhaps alongside “What’s the next best action?” we need to get better at asking “What does this relationship need from us right now?” Sometimes the answer will absolutely be a thoughtful piece of content, a personal conversation, an invitation, a stewardship touch, an ask, or another clearly defined next step. Sometimes the answer may simply be time.

As AI and automation make it easier for us to act on every signal, I suspect that distinction will become more important, not less. The most audience-first organizations won’t just become exceptionally good at knowing what to do next. They’ll develop the judgment to recognize when doing something isn’t actually better for the relationship. Maybe one of the most sophisticated workflows we can build really does end with Next Best Action: Nothing Happens. Not because the system failed to act, but because it understood that, for now, nothing needed to happen.

What do you think? Let’s Talk!