# Scaling as an investment decision Why engineering visibility becomes critical as companies scale. ![](https://framerusercontent.com/images/aHHJTe8rMRUw3vhronp54jURux4.jpeg?width=800&height=800) Bernardo Hernández Co-founder May 20, 2026 Scaling taught me more than strategy ever could. Not because strategy isn’t important, but because scaling forces you to confront reality. It exposes what actually works inside an organization and what only sounded good in a plan. When you are small, you operate as a hunter. You do things yourself, you move fast, and the feedback loop between action and outcome is immediate. As you grow, that changes completely. Your role shifts from doing to enabling, from executing to creating the conditions for others to perform, and that transition is where most of the real lessons happen. But there is one idea that becomes clearer over time: Scaling is not something that happens to you, it is something you choose. ## **From growth to investment** Most companies treat scaling as a natural response to growth. Demand increases, the roadmap expands, and the instinct is to add more people to keep up. It feels logical, almost inevitable. But as I’ve said before, scaling is not a consequence of growth, it is an investment decision. Every time you add headcount, you are allocating capital with the expectation of a return, not just in output, but in impact. And like any investment, it should be evaluated against alternatives. The problem is that it rarely is. ## **What scaling actually introduces** When you scale a team, you are not just increasing capacity, you are also increasing complexity. I’ve been there managing teams of hundreds of people, I know what it means: More people means more coordination, more communication paths, more dependencies, and more distance between decisions and execution. In the early stages, this trade-off is favorable. As the system grows, it becomes less predictable. The marginal return of each additional hire decreases, and in some cases, turns negative. At some point, adding more people does not accelerate the system anymore, it can become even slower. This is where scaling stops being an obvious decision and becomes a question of return. ## **The moment where it breaks** In the last years I have sat in dozens of board meetings across different companies, and there is a moment that repeats itself almost every time. Engineering presents a coherent story. Progress is shown, hiring plans are outlined, AI adoption is mentioned. For a moment, it feels like we understand what is happening. Then the questions start. - *Are we actually shipping faster than before?* - *Are we getting a good return on what we are investing?* - *How do we compare to others?* - *Is AI improving productivity or just changing how work is done?* - *Did costs scale responsibly?* - *Do we have the best people?* - *Is everyone contributing at the level we expect?* These are not complex questions. They are the most natural questions you should ask when a function represents 30 to 40 percent of your spend. And yet, in most cases, they don’t have clear answers. Actually, I wrote a [blog about it.](https://pensero.ai/blog/board-meetings-and-the-engineering-blind-spot) ## **Why scaling makes this harder** When organizations are small, you can operate without answering these questions precisely. You rely on proximity, intuition, and trust. You can feel whether things are working. As you scale, that breaks: You are making larger investments, with longer feedback loops, and with more layers between decision and execution. **When scaling, the cost of being wrong increases, and the ability to rely on intuition decreases.** Scaling amplifies everything. If the system works, it scales performance. If it doesn’t, it scales inefficiency. And without a clear way to measure what is happening, you cannot tell the difference. ## **The visibility gap** This is where most organizations struggle. In functions like sales, visibility is built into the system. You can see pipeline, conversion, revenue, and performance with clarity. That allows you to evaluate whether scaling is working. Engineering has historically been different. We rely on operational metrics that help teams organize work, but do not translate into value at a leadership level. So the conversation becomes interpretative. Each person builds their own version of reality, and decisions are made on narrative rather than evidence. When you combine that with scaling, you are effectively increasing investment without increasing understanding. That is a fragile position to be in. ## **Scaling in the age of AI** AI makes this even more critical. It is increasing the output that a single engineer can generate, but it is also increasing variability across teams. Some teams are compounding their performance. Others are not. The difference is no longer marginal. In that context, **scaling blindly becomes even more risky.** If productivity per engineer is increasing, the question is not how fast you can grow the team, but whether scaling is the highest-return decision available. In many cases, improving how the system operates will generate more value than expanding it. But you cannot make that decision without visibility. ## **A different kind of discipline** Treating scaling as an investment changes the conversation. It forces you to define what you expect to gain before you commit resources. It forces you to measure what actually happens after. And it creates accountability around return. It also forces you to answer the questions that boards are already asking, but that most organizations cannot answer with confidence. - *Are we getting better?* - *Are we getting a return?* - *Are we improving faster than others?* - *Is AI helping or just adding cost?* These questions are the foundation of how you decide to scale and to do it in a way that actually increases performance.