the cost of choosing what to build

there is a question that keeps coming up when i talk to other builders.

not the obvious ones. not "what are you working on" or "how is traction." it is a quieter question. one that usually surfaces after the first few months when the initial excitement has settled and the real weight of the work becomes clear.

why this, and not something else.

it sounds philosophical. maybe even self-indulgent. but i think it is actually one of the most practical questions a small team can ask itself. because focus is not free. every system you decide to build is a system you decided not to build. every problem you go deep on is a problem you walked away from.

at SAGEA, we have been sitting with this question a lot lately.

the branching problem

early on, the tempting move is to branch.

you build one thing. it works reasonably well. you notice five adjacent problems that your system could plausibly address. the instinct is to expand. capture more surface area. build more things. cover more ground. this instinct is almost always wrong, at least early.

the thing that kills small teams is not ambition. it is diffusion. spreading energy across too many branches before any single branch has gone deep enough to matter.

what we have been trying to do instead is go narrower and harder on the things we have already committed to. not because expansion is bad, but because depth produces something that breadth cannot. it produces genuine understanding of where the problems actually are. and that understanding is the only thing worth building on top of.

what research actually looks like

people have a stylized image of research. whiteboards. papers. late nights of intellectual exploration with no particular deadline.

the reality, at least for us, looks very different. research mostly looks like running into a wall. spending two weeks on an assumption that turns out to be wrong. building something that works perfectly in a controlled environment and completely falls apart when you introduce real inputs.

but those failures are the work. not obstacles to the work.

i have been thinking about this recently because some of the projects we have been deepest on started as pure research exercises. no product goal. no immediate application. just sustained curiosity about a class of problem. what is interesting is that the things we understand most deeply right now, the things that feel closest to becoming something real, are almost always the ones that started that way. not from a product spec, but from a genuine question that kept nagging at us.

the cost of conviction

there is a cost to conviction that does not get talked about enough.

when you believe something is the right direction, you have to be willing to stay in it even when it is not producing visible results. that is genuinely uncomfortable. especially in an environment where everything is moving quickly and the pressure to show output is constant.

but i think the teams that last are the ones that learned to sit in that discomfort without abandoning the direction. not stubbornness for its own sake. if the evidence changes, you update. but there is a difference between updating on evidence and abandoning a direction because progress is slow.

slow progress on the right problem is still progress. fast progress on the wrong problem is just expensive distraction.

where we are

we are in a phase right now where a few things we have been quietly building are starting to crystallize.

not finished. not ready to talk about in detail. but the shape of what they are is becoming clearer. and there is a certain feeling that comes with that clarity. not excitement exactly. something quieter. a sense that the direction was right and the depth was worth it. i have written before about how compounding works in systems. this is the part of the curve that precedes it. the part where the work still looks invisible from the outside but the foundation is starting to feel solid from the inside.

that is roughly where things stand.

we keep building.