Direct answer
People don't always ask for a brand, or even a product: they describe a problem. "How do I remove a grease stain," "what can I do about aphids," "my back hurts." AI answers by suggesting solutions — and sometimes types of products, even brands. Knowing whether AI cites your product category as an answer to these everyday problems means capturing demand upstream of the brand, at the exact moment of the need. For many everyday products, that's where the bulk of it is decided.
The problem
You may think of your AI visibility in terms of "brand" or "product category." But a large share of your potential customers think of neither: they have a problem to solve, and they describe it to AI just as it is.
Someone asking "how do I remove a grease stain from a garment" isn't looking for your brand of stain remover. They aren't even looking for "stain remover." They're describing their problem. And AI's answer — citing this type of product, this tip, this brand — decides whether you exist at that moment. If AI doesn't mention your type of solution, you're absent from the need before the question of choice even comes up.
The idea to grasp
There's a level of search upstream of the brand and even of the product: the problem. The typical journey looks like this:
1. The problem: "my back hurts" → AI suggests leads (stretches, seeing a professional, accessories…). 2. The category: if AI mentions a type of product ("a lumbar cushion can help"), your category enters the conversation. 3. The brand: if AI goes as far as naming brands, the choice gets more specific.
Many brands measure only level 3 (am I cited when brands are named). But if you don't exist at level 1-2, you never get to play at level 3: the customer never reaches you.
What gets measured here:
- Is your type of product cited as a solution to a given problem? ("for this stain, an enzymatic stain remover" → your category exists).
- On which problems do you appear, and on which are you absent? A map of the needs that lead — or don't lead — to you.
- At what point in the answer: does AI suggest your type of solution right away, or only as a last resort?
This is especially decisive for everyday products: household care, drugstore items, DIY, gardening, wellness. Fields where people massively describe problems to AI rather than searching for a brand.
What you hear everywhere
"My customers know my product, they search for it directly." Some, yes. But many start from the problem, not the product. You only capture those if AI cites your type of solution at the right moment.
"If my product is effective, AI will suggest it." Real effectiveness and presence in AI's answers are two different things. AI suggests what is associated with the problem in its data — not necessarily the objectively best solution.
"Measuring that is too granular." On the contrary, it's often the biggest pool: capturing the need upstream reaches a volume far greater than brand searches alone.
My stance: facts only. Assuming AI recommends your type of product in response to a problem is a hypothesis. Measuring it across a panel of real problems is data — and often a surprise.
My vision: measure presence at the problem level
From here on, the register changes: we describe the instrument.
Measuring presence upstream of the brand means:
- Building a panel of real problems your customers describe (worded the way they word them).
- Querying the AIs on these problems, repeatedly, to see whether and how your type of solution emerges.
- Measuring: whether your category is cited, how often, and where in the answer.
- Mapping the problems that lead to you and those that ignore you.
- Comparing modes and providers, dating and sealing.