What this method produces. An honest read of the one public quantitative instrument every student has: search interest as a signal — with its four traps disarmed.
Trap 1 — The scale is relative, not volume. 100 means "the peak of interest for THIS query in THIS period and region" — not "a lot of searches". A tiny niche term and a giant category both peak at 100. Consequence: never read a number off one lonely curve.
Step 1 — Calibrate by comparison. Add a reference term whose scale you understand (your category's generic name, an established competitor product). "Balaclava" against "scarf" tells you the real order of magnitude; "balaclava" alone tells you nothing.
Trap 2 — Seasonality wears a trend costume. "Linen dress" surges every May — that is summer, not a movement. Step 2 — Compare year over year, not month over month: is THIS May higher than the last three Mays? Growth is measured against the same season, never against the previous month. Use 5-year view minimum for any trend claim; a 90-day window can only show spikes.
Trap 3 — Interest is not sales. Search measures curiosity: scandals, memes and film releases all spike it without selling a single unit. Step 3 — Before quoting a spike, name its likely driver (check the dates against news) and say what would confirm demand: sell-through, resale activity, new drops.
Trap 4 — Geography is a filter, not a footnote. A worldwide curve averages away the story. Step 4 — Read your target market's curve separately; a signal exploding in Korea and flat in France is a Korean signal for now — note it as spread potential, not local demand.
Step 5 — Write the finding in signal grammar. Not "the trend is growing" but: "search interest for X in [region]: third consecutive year-over-year rise in the same season; calibrated against Y it reached ~40% of an established category; driver check shows no one-off event. Interest signal: medium-strong. Demand: unconfirmed — needs sell-through."
Common mistakes. Quoting the 0-100 number as market size. Comparing terms in different languages as if they were one market. Claiming a trend from a 90-day window. Treating a film-driven spike as organic demand without naming the film.