The Relational Stack: Beyond Product-Centered Tech

Malena Data Ernani is one of nine rising community leaders who took part in Aspen Digital’s Community & Tech Leaders Initiative. Over the course of a year, the cohort. deepened their understanding of emerging technologies and explored how community perspectives could shape the innovation ecosystem. This blogpost and accompanying resource was produced at the conclusion of the Initiative. 

We routinely fund, measure, and celebrate technological interventions while treating the human intervention that makes them work as an externality. A platform gets credit for increasing access to public benefits, while the promotoras who build trust and help people navigate the process are treated as supplemental. A mental health app is purchased to support students, while the relationships, school policies, community conditions, and institutional coordination shaping their well-being sit outside the intervention. An AI tool can accurately identify bias, but is someone accountable for whether that insight travels through communities and institutions in a way that produces systemic change. The “inclusive” technology we build may work exactly as intended and still leave the larger human system fundamentally unchanged. 

I saw versions of this tension repeatedly throughout my experience in the Community and Tech Leaders cohort. At a moment when AI and other emerging technologies are dramatically expanding what we can build and how quickly we can build it, we have become increasingly sophisticated at making technology 

more inclusive, accessible, accurate, and responsive to the people who use it. Product Equity and participatory design have pushed the field forward by asking who is represented, who participates, and who benefits from a new technological intervention. But there is a risk in allowing the product to remain our primary unit of success. High engagement does not necessarily mean greater belonging. An accessible public-benefits platform does not necessarily mean people can navigate fragmented agencies. An accurate AI system does not guarantee that anyone can act on what it produces. A technology can succeed at the product layer while failing to produce meaningful change across what I call the Relational Stack. 

The Relational Stack is a way of seeing technology not as the system itself, but as one layer within the human systems that make change possible. We already use the idea of a technology stack (or tech stack for short) to describe the collection of tools and technical layers that make our work possible (think Zoom, Slack, Gmail, and the infrastructure connecting them). The Relational Stack, however, asks us to zoom out and consider the human systems those technologies actually live within: relationships that create trust, belonging, and connection; communities that organize around shared needs; institutions that shape how we access care, education, services, and opportunity; technology itself; and infrastructure, including policies, processes, interoperability, and coordination capacity, that allows these layers to work together. These Relational Stack layers are not separate; they constantly shape one another. The point is not simply that they coexist. Human outcomes are produced across the stack, and our interventions should be designed, funded, and evaluated accordingly. The framework therefore encourages us to center the human outcome (not the product) first: What are we trying to achieve? What systems are already responsible for that outcome? Where are they failing to coordinate? And where, if anywhere, can technology strengthen their capacity? 

This changes where we start (and where we end up) when building technological interventions. Product Equity has helped move community voice upstream, asking who is represented, who participates, and whose needs shape a product from the beginning. But even when community input meaningfully shapes the technology, the feedback loop can still close around the product itself: community needs → product design → community feedback → product iteration → improved product. The Relational Stack asks us to keep that loop open beyond the technology. Consider a school facing rising student mental-health

challenges. A product-centered approach might ask which mental-health technology to procure and then measure adoption or engagement. A Relational Stack approach starts with the shared outcome: What would it take for students to actually be better supported across the stack? It asks what is contributing to the problem across relationships, community conditions, school practices, access to care, policy, and infrastructure, and only then where technology might strengthen that response. The same is true for someone navigating disconnected healthcare, transportation, and social-service systems. Instead of asking “What app can we build?” we might ask: “How do we prevent this person from having to coordinate five fragmented systems on their own?” Perhaps technology can connect appointments and transportation, surface disruptions, or make institutions more interoperable. Perhaps the missing intervention is policy, funding, trust-building, or simply better coordination among humans. The sequence matters: human problem → shared outcome → human system → coordination → technology. And the learning should flow back into the whole system—not just into the next product iteration. 

This shift also forces us to confront incentives. We have strong incentives to fund what can be built, measured, monetized, and scaled. Relational work (building trust, connecting institutions, sustaining community networks, coordinating care) is often distributed, difficult to attribute to a single actor, and consequently easier to underfund or leave invisible. We increasingly use technology to compensate for our failure to coordinate as humans, and then build more technology to compensate for the fragmentation created by that technology. Even when a technology addresses a genuine need, funding the product without funding the relationships and coordination required to make it effective leaves us with an incomplete intervention. If we want technology to advance human progress, we cannot keep funding the technological intervention while treating the human intervention as an externality. 

The accompanying two-page visual illustrates the logic of the Relational Stack and how it can help us think beyond product-centered technology. I hope it can serve as a shared language for technologists, funders, public-sector leaders, and community practitioners to ask not only Is this technology inclusive? or Does this product work?, but: What shared human outcome are we accountable to? What does achieving it require across the stack? What are we funding and measuring? And where does technology actually belong? Technology should not become the operating system that everyone else is forced to organize around. It should be an integrated part of our human operating system—an extension of our collective capacity to connect, coordinate, care, and act. 

Ultimately, the Relational Stack is a challenge to rethink what we mean by technological progress. The goal is not to build technology that operates alongside human systems, but to build technology that strengthens our capacity to connect, coordinate, care, and act together. That means measuring progress not only by what we can build, how many people use it, or how efficiently it scales, but by what becomes more possible for people, communities, and institutions because it exists. Technology is at its most powerful when it becomes part of the infrastructure of human progress, not a substitute for it or an addendum to it. 

What kinds of human lives can technology help make possible and, most important, help stitch together?

The views represented herein are those of the author(s) and do not necessarily reflect the views of the Aspen Institute, its programs, staff, volunteers, participants, or its trustees.