There is a difference between the kind of discovery that happens when you stumble on something in the right moment, and the kind that happens when you are looking. Researchers from the University of Pennsylvania and American University have mapped this distinction empirically, identifying two distinct patterns of curiosity: hunters, who connect closely related ideas to fill specific knowledge gaps; and busybodies, who follow associations across wider terrain, forming loosely connected but expansive networks. Both patterns are active. Both produce value. But they depend on different conditions.
The problem most organisations face is not a shortage of curious people. It is a shortage of the conditions in which curiosity produces something lasting.
Where knowledge goes when the meeting ends
Most platforms for collective thinking are designed to produce an output and close. The workshop ends. The consultation closes. The document is written. The understanding that was building — the connections forming between people who had not previously been in contact, the ideas gaining traction at the edge of the discussion — expires with the event that produced it.
This is not a small problem. The most valuable thing that emerges from collective thinking is often not the conclusion. It is the unexpected connection: the moment when a perspective from one domain illuminates something in a completely different one, and a new understanding becomes possible that neither contributor could have reached alone. These moments are not plannable. They arise from the structure of the encounter — who is in contact with whom, what kinds of knowing are in the space together, how long the deliberation continues.
Serendipity is not random on a well-designed platform. It is the predictable result of sustained contact between epistemically diverse perspectives.
What assisted serendipity means
Assisted serendipity is not algorithmic recommendation. It is the structural design of conditions that make unexpected connection more likely — and the preservation of those connections when they form.
On Hunome, a SparkMap does not close when the conversation reaches a conclusion. It continues to develop as new contributors join, as existing contributors add to what they previously built, and as the connections between contributions become visible over time. A perspective contributed in month one may not connect visibly to another contribution until month three, when a new contributor builds the bridge. The platform preserves both perspectives and the connection between them — not as an archive, but as a living structure that can be developed further.
The Ignite characterisation architecture makes this possible by giving every contribution its epistemic type. When a contributor with lived experience builds on a contribution grounded in research, the connection is not just between two pieces of content — it is between two ways of knowing. That kind of connection does not happen in a survey, where everything becomes a number. It does not happen in a workshop, where the conversation moves on. It happens in a space designed to hold epistemic diversity in sustained contact.
What this produces
The clusters that emerge from a SparkMap are not specified in any brief. They emerge from the structure of the collective's own thinking, developing in relation to itself. In the demographics deliberation that Hunome and Futurely ran in 2024, seven distinct clusters emerged that no research brief had anticipated. Several of them formed from connections between contributions that were made weeks apart, by contributors who had never met.
This is assisted serendipity: not luck, but structure. Not algorithms optimising for engagement, but architecture designed to sustain the conditions in which unexpected understanding becomes possible — and to preserve it when it forms.
For groups that return to a shared question over time, this compounds. The aha moments that Kierkegaard describes as the crystallisation of understanding — the moment when something that was obscure becomes suddenly clear — do not happen on demand. But the conditions that make them possible can be built.