Design is a business decision, not a coat of paint

Every few months a founder tells us the same story: they paid for a redesign, the product looks better, and nothing changed. Sign-ups flat. Retention flat. The team is proud of the screenshots and quietly disappointed in the numbers.

The problem is almost never the designer’s taste. It is that the project was framed as “make it look better” instead of “make something specific happen.” Looking better is a side effect of good design, not its goal.

Before any design work starts at Pixi, we force one question: which number should this change? Activation rate, support tickets, time to first value, checkout completion — anything, as long as it is measurable and honest. That single constraint changes everything downstream: what we research, what we prototype, what we test, and how we know when we are done.

It also changes the conversation with stakeholders. Taste is subjective and endless; metrics end debates. When someone wants the logo bigger, the question is no longer whose opinion wins — it is whether a bigger logo moves the number we agreed on.

So before your next design project, write one sentence: “This project succeeds if ___ goes from ___ to ___.” If you cannot fill in the blanks, the project is not ready to start. When you can, design stops being a cost center and starts being the most leveraged investment in the company.