There is a version of systems thinking that works well as a professional credential. You put it in a LinkedIn summary, it signals that you think holistically, see relationships, and work across silos.
Then there is the version that shows up when you are actually building a product that helps people map a system, and you realize the way teams talk about systems is cleaner than the way they understand them.
Building Gestalt taught me the second version.
What I thought I was building
Gestalt started from a simple frustration: every systems conversation I had been part of, whether in consulting work, product teams, or architecture discussions, began with a blank surface and ended with a map that nobody opened again.
The map represented something real: the dependencies, the relationships, the flow of decisions and data through an organization. It captured a moment of shared understanding, then became a static artifact, and the next time the same problem came up, the conversation started again from scratch.
I thought the problem was persistence. If the map lived somewhere accessible and was maintained rather than produced for one session, the value would compound. The team could build on the last version instead of reconstructing it.
So I built a persistent visual system mapping tool: portals that connect, relationships that stay visible, nested maps that can be updated incrementally rather than replaced wholesale.
That was still the right product direction. What I got wrong was my assumption about what people would put inside it.
What building it revealed
I assumed teams would map systems the way systems thinkers describe them: components, relationships, flows, dependencies. Clean, structured, compositional.
What I observed in my own use was messier and more useful. People do not start by mapping the system as it is. They start by mapping the part they are anxious about, the dependency they do not trust, the connection they half-understand, or the question they keep coming back to.
That does not make the map wrong; it makes it honest.
A clean diagram can be useful when the system is already understood, but a messy map often tells you more about the team's real state of knowledge. It shows where the structure is known, where it is inferred, where the relationship is suspected, and where nobody has enough confidence to draw the line yet.
That changed how I thought about the product. Gestalt cannot only support polished representations. It has to support the incomplete one: the map where someone can say, "I know this connects to something, but I do not know what yet." If the tool forces premature structure, it becomes another diagramming surface. If it lets the uncertainty stay visible, it becomes closer to a thinking tool.
What this means for product work
Systems thinking as a credential often assumes an almost omniscient view: see the whole, map the relationships, reason about the emergent behavior.
Systems thinking as a product discipline is more grounded than that. It is the discipline of building representations that are honest about what the team does and does not know. The point is not to look complete; the point is to make the incompleteness useful.
The most valuable thing a system map can do is show where understanding breaks down: the dependency everyone mentions but nobody owns, the relationship that exists but has never been mapped, the assumption that is load-bearing but invisible, the part of the workflow that lives entirely in one person's head.
A map that exposes those gaps is more useful than a map that performs confidence.
That is what Gestalt is trying to be: not a whiteboard replacement for clean diagrams, but a system map that helps teams make their understanding legible, including the parts they have not figured out yet.
Whether it fully achieves that is still an active question. The product is live at getgestalt.app. The current public pricing shows a free evaluation tier and Pro at $19/month.
What building it has clarified for me is that the value of systems thinking in product work is not having the complete picture; it is knowing precisely where your picture is incomplete.
That is what drives better questions, and in product work, better questions are often more valuable than confident answers.
Built by Moe Hachem. mghachem.com