Mayja Match.

When Mayja Skin came to us, the ask sounded like a shop. Let people buy at the booth. We spent the first conversation working out that selling was not the constraint. Deciding was.

A fragrance booth converts on a conversation, and a conversation does not scale. The person behind the table can only talk to one customer at a time, and the good version of that talk — what do you normally wear, do you like sweet or clean, here, smell this — takes minutes. Multiply that by a three-day festival and the line becomes the ceiling. Everyone past a certain point in the queue walks.

So we built the conversation, not the shop. Two entry points, because the two kinds of customer need different first questions. Six scent families written the way customers talk rather than the way perfumers do. A mood step capped at two taps, because a third tap is where someone standing in a crowd gives up. Then three named scents she can point at.

The two decisions we would defend hardest are the small ones. The first is that the layering suggestion lives on the match card rather than at checkout — the moment a customer is pleased with a recommendation is the only moment the balm sells itself, and putting it a screen later would have wasted it. The second is the list signup. Every booth at every festival collects emails on a clipboard and loses half of them to handwriting. We made it a QR that opens the customer’s own messaging app with the text already written. She does not type anything. She presses send, and she is on the list from her real number.

Checkout takes whatever the crowd already uses rather than whatever we would have preferred to integrate. At a festival that matters more than elegance.

We handed it over before the doors opened. Across the weekend Mayja’s own team ran it — no one from Isaac and Co. stood at that booth.