Ideas that help people decide and build better.

Technology Starts Before Code

Before choosing a technology, understand the problem, the context, and how the decision will affect the operation.

Choosing technology often looks like a technical decision. Which language should we use? Which database? Build or buy? What architecture will support the next stage?

These questions matter, but they are rarely the first ones. Before them sits a harder question: what problem needs to be solved, and what changes in the operation when it is solved?

The problem before the solution

A request for software usually arrives with an imagined solution attached. It may be an application, an integration, an automation, or a system replacement. The proposal may already have a name, a deadline, and sometimes even a chosen technology.

The most important work at that point is to separate the problem from the suggested solution. What is happening today? Who is affected? Where does the operation lose time, information, or decision-making capacity? What needs to be different afterward?

This is not bureaucracy. It is a way to avoid building the right thing for the wrong question.

Context is part of architecture

The same problem can require very different decisions depending on its context. A small operation may benefit from a simple, well-configured tool. A distributed operation may need integration, traceability, and clear rules of responsibility.

There are also constraints that do not appear in a diagram: manual processes, knowledge concentrated in a few people, legacy systems, commercial dependencies, security requirements, and maintenance capacity.

Ignoring these conditions produces an architecture that looks elegant on paper and is difficult to sustain in daily work. A solution’s architecture includes the environment where it will be used.

Operational impact

A technical decision does not end when the code is delivered. It changes routines, responsibilities, support costs, and the way people experience the work.

It is therefore worth asking:

  • Which activity will be reduced, removed, or created?
  • What information will exist, and who will be able to act on it?
  • What happens when an integration fails?
  • Who maintains the solution after the first delivery?
  • How does the operation identify that something is outside the expected path?

These questions bring technology closer to the business without reducing technology to a return spreadsheet. They make visible effects that often appear late, when change has already become expensive.

The decision may be not to build

Understanding the problem better may lead to the conclusion that it is not time to build. A process rule may address the cause. An existing configuration may be enough. The problem may still be unclear and need observation before a permanent solution is chosen.

Not building does not mean ignoring the opportunity. It means preserving the ability to decide when there is enough information.

Likewise, choosing a simple solution is not a lack of ambition. When it fits the context, simplicity reduces dependencies and leaves room to learn from real use.

Technology as a business decision

Once the problem is clear, the technical choice becomes more precise. The team can evaluate alternatives by what they allow people to do, maintain, and change — not only by what is currently visible in the market.

That evaluation considers the solution’s whole life: deployment, operation, observability, security, evolution, and eventual replacement. Code is an important part, but it is not the whole product of the decision.

Technology starts before code because the quality of a solution depends on the quality of the question. Understanding context, making impact explicit, and choosing responsibly is what turns a technical decision into useful execution.