Why a second category is harder than a second city

From outside both look like “plus one unit”. In practice they load the company differently, and confusing them is expensive.

Copying versus inventing

Opening a location in a new city transfers an already described process: shift standard, training, supply, quality control. The main work is logistics and people. Opening a new category means writing a new standard, finding new suppliers, learning to measure a different kind of quality and explaining a new ritual to the guest — all at the same time.

The first scales discipline. The second stress-tests it, and usually finds the places where there was no discipline, only personal control.

Where the shared layer breaks

The point of the system is that categories use a shared customer and operating layer. But a shared layer is not universal by default: quality metrics for coffee and for food do not reduce to each other, supply cycles differ, and the visit context overlaps only partially.

So the question before entering a category is not “can we sell this”, but “what exactly from the working layer applies here without being rewritten”.

Four entry criteria

• The category increases the frequency of useful contact, not the number of signboards.
• It uses the customer layer that already exists instead of requiring a second one alongside.
• It strengthens the operating system rather than duplicating it with another team and other rules.
• It does not break the discipline of the existing business: attention to the core operation does not drop.

If even one criterion fails, the category stays in roadmap or vision status. That is not a refusal — it is a refusal to announce a launch before it is justified.

Why we do not name dates

Entry timing depends on when the shared layer survives testing, not on a calendar. A named date would turn a roadmap into a promise, and we keep those statuses apart on purpose: a pilot is not a product, a roadmap is not a result, a vision is not an obligation.