Dinner’s Sorted. Who Picked the Sauce?

*On Clementine, and brands that might never make the shortlist.*

At seven in the evening, humanity is not always in the mood to exercise freedom of choice. Sometimes it’s standing in front of the fridge, hoping someone will finally tell it what to eat.

Against that backdrop, Clementine, Instacart’s AI shopping assistant, makes perfect sense. It helps people plan meals, find recipes and put together a shopping list through conversation. Fewer tabs, fewer comparisons, less chance of eating cheese for dinner in front of an open fridge. How Clementine works — Instacart.

But somewhere between “I fancy pasta” and the finished list, a question emerges that should interest brands: why this pasta and this sauce?

The final decision may still belong to the shopper. By that point, though, somebody has already narrowed down the options. If your product isn’t among them, even the most persuasive label won’t help. Nobody will see it.

For us, this is a reason to look at how a brand describes its product beyond the beautiful mock-up. “A premium natural product, made with love.” Touching. Is it spicy? What’s in it? How much is in the jar? What does it go with? Love on the ingredients list answers precisely none of these questions.

People already need clear product information. In a digital catalogue, it also gives systems specific data to process. That isn’t a magic formula for getting an AI recommendation. But a product whose only available attribute is “special” is making introductions unnecessarily difficult.

For a small brand, several outcomes are possible. An assistant could identify it as a precise match for a request and introduce it to a new customer. Or it could keep suggesting familiar products, leaving discovery for another evening. We aren’t claiming either outcome for Clementine in advance. We’re interested in the kind of shopping experience these systems will eventually create.

Convenience is easy to measure: dinner decided, list ready. What’s harder to notice is everything that never made the list.

And while brands work out how to win people over, another task is taking shape: getting the chance to be considered at all. Preferably before the shopper accepts the recommended sauce and closes the app.