Working out how many recipes a sauce for professional kitchens needs

The question

A food company’s sauce for professional kitchens had grown apart from market to market, shaped by local taste, local rules and local production.

The company wanted to know what that variety was worth to the chefs who use it: what makes the best sauce for their kitchens, in each market and across them, and whether one of its existing recipes could serve them all.

Our approach

Chefs defined the ideal before seeing any product, naming the qualities that matter and how they relate, so each recipe was judged against criteria they had set.

Products came blind-coded for chefs to rank quality by quality, each chef built a dish with the products at either end of their ranking, and chefs said how they would prove a quality was there, so the judgment happened as it would in a working kitchen.

Local moderators worked from one adaptable plan, listening closely to the words chefs used in each language, so the markets could be compared fairly.

What the team came away with

For every quality in every market, the product team had a sweet spot and the range chefs would accept around it.

A difference between recipes could now be read against what chefs would tolerate, and design principles turned that map into guidance the developers could build recipes from. Because chefs had set the criteria before tasting, whether one recipe could serve every market became a question the company could settle against their standards.

The chefs’ own words for each quality came with it, in every language, so local teams and developers could talk about the sauce the same way.