Method
Importance weight
An importance weight expresses how much a tag, category or assistant should matter to a brand's overall score. Weights redistribute influence between parts of the measurement; they never add to it. A heavily weighted tag moves the headline more, but no amount of weighting can push a score past what the answers support.
Importance, never volume
Weights are set from how much a part of the buying conversation matters to the brand, not from how many prompts happen to sit in it. A tag holding eight prompts can be weighted above one holding thirty, if eight is where the buyers are.
What they cost
Unequal weights mean a sample carries less information than its row count suggests — the effective sample size is smaller than the number of answers collected. That is a real trade, and it is why the confidence range on a heavily weighted slice is wider than a naive count would imply.
Eight prompts that matter more than thirty
A tag holding eight questions your buyers genuinely ask can be weighted above one holding thirty peripheral ones. The overall score then moves more when the eight move. What the weights cannot do is invent performance: every input is already a share of answers, so no weighting makes a score higher than the answers support.
Related terms
- Effective sample size — The price of unequal weights, paid in statistical strength.
- Prompt-mean rule — Frequency of sampling is deliberately not a weight.
- Confidence range — Wider on heavily weighted slices, for that reason.
- Prompt taxonomy — The tags and categories the weights are applied to.
Last reviewed . Definitions are reviewed quarterly, and whenever the underlying measurement changes.
Don't let AI decide your brand's future without you.
See exactly where you stand vs. competitors—and what to do about it.