Liking is evidence. It is not adoption.

A single score can hide several different events

Detection, perception, evaluation, preference, choice, trial, continued use, switching and adoption describe different events. Collapsing them creates evidence that sounds more decisive than it is.

A forced paired-preference task identifies which of two options was selected under the task. It does not establish that the selected option was liked. Both options may be disliked. A liking rating is evaluative, but it is not a choice made under price, access, identity, habit, social meaning or competing alternatives.

Reward research also distinguishes hedonic “liking” from motivational “wanting” and learning. Those components can interact, but they are not interchangeable (Berridge et al., 2009). Product adoption adds still more layers: availability, friction, routine fit, trust, cost, cues and the alternatives a person is expected to replace.

A useful ladder is: detect → perceive → evaluate → prefer → choose → trial → continue → switch or adopt.

Each transition can fail for a different reason

A person may notice the intended feature but interpret it negatively. They may like the experience but find the product inconvenient. They may try it but never integrate it into a routine. They may continue using it without switching away from the alternative the organization expected it to replace.

That is why food choice and product behaviour are rarely explained by a single attitude or rating. Past behaviour, habit, context and memory can be more predictive than stated intention alone (Köster, 2009).

First use is a special condition

Attention is concentrated, novelty is high, instructions are fresh and the research team often controls serving, sequence and task. Repeated use introduces learning, adaptation, maintenance, cost, social context and habit. The product must survive a different set of constraints.

This is not an argument against liking

Liking is useful evidence about evaluation under defined conditions. It is not a universal proxy for what people will do next. Sensory and consumer science has long argued for measuring beyond liking when the decision concerns behaviour, ritual, context or continued use (Meiselman, 2013).

The research question should name the failure point the product must survive. If the risk is perception, test perception. If the risk is immediate acceptance, test that. If the decision concerns continued use, switching or adoption, the evidence plan needs behavioural outcomes measured over time.

Liking is evidence. It is not adoption.