Insight

An interpretation of a theme — what the cluster of signals means, why it exists, and what should be done about it. The step that turns synthesis into a decision.

Signal, theme, and insight are three different things, and conflating them is the most common mistake in feedback work. A signal is a raw piece of input. A theme is a cluster of signals that share an underlying need. An insight is the interpretation of the theme: what the pattern means about the customer, why it exists, and what the team should do about it.

An insight is not a feature request. "Customers want CSV export" is a theme summary. The insight is something like: "Our customers move data into spreadsheet-based analysis weekly, and the absence of CSV export forces them into screenshots — we should treat external analysis as a first-class workflow, not a leaky edge case." The first sentence is observation; the second is a strategic frame the team can act against.

Most automated feedback tools stop at themes. The leap to insights still belongs to humans: a PM (or a PM-and-LLM together) reads the cluster, looks at the underlying context, and writes the interpretation. The quality of a roadmap is gated by the quality of insights, not the quantity of themes.