Trends — launching early enough to ride, late enough to avoid the cliff.
A trend is a temporary spike in demand. Catch it early and you're alone on the SERP at the moment a wave of shoppers arrives. Catch it late and you import inventory into a saturated category that's about to collapse. The skill is reading the curve, not predicting it.

Trends are the most-discussed and least-disciplined angle in product selection. Every seller has chased one. Most lose money because they confuse a fad for a trend, or because they launch at the wrong point on the curve. This episode is about the curve itself — what shape distinguishes a real trend, where on it a launch makes sense, and where it doesn't.
The curve
A real trend has four phases:
- Emergence. A handful of early signals — niche social posts, a small subreddit, a few imports. Amazon search volume is low but accelerating. The SERP is mostly empty or filled with adjacent products.
- Acceleration. Search volume doubles month over month. The first dedicated listings appear. Review counts on the early movers grow visibly week over week.
- Saturation. Hundreds of listings appear. Sponsored slots fill. Search volume keeps rising but bid prices rise faster. Margin compression begins.
- Decline. Search volume rolls over. Inventory bought in saturation arrives into a falling market. Sellers cut price, the category becomes a loss-cutting exercise.
Where to launch
The window is the late emergence / early acceleration boundary. Search volume is unmistakably climbing, sourcing is still possible at reasonable lead times, and the SERP still has empty slots. A launch here arrives in time for the acceleration phase, when buyers outnumber sellers. Launching earlier risks a fad that never reaches acceleration. Launching during saturation is a losing trade.
Distinguishing trend from fad
A trend has a structural reason that explains why demand is rising — a new use case, a price-point shift, a regulatory change, a demographic moment. A fad has only social-proof momentum. Three diagnostic questions:
- Does the product solve a problem that existed before the trend? If yes, the trend can persist.
- Does Google search data corroborate? Amazon-only rising volume with flat Google is more often a fad.
- Are adjacent, durable categories also lifting? A trend in a single keyword that doesn't touch its neighbours is usually a fad.
Inventory discipline
Trends fail seller-side most often on inventory. The temptation is to order large quantities once acceleration is obvious, to capture the wave. The discipline is the opposite: order small initial quantities at faster turnaround, re-order weekly based on actual sell-through, and stop ordering at the first sign of saturation — not at the first sign of decline. By the time decline is visible, the inventory ordered against saturation is already on a container.
The exit
Every trend ends. The decision isn't whether to exit, it's how. Two clean exits work: liquidate aggressively in early decline (price drops, expanded sponsored, accept lower margin to clear), or pivot the listing to an adjacent durable category if the product has a non-trend use case. Holding inventory through the trough hoping for a second wave is the most expensive choice on Amazon.
Watch Module 7 · Episode 06 — Trends. (German)
The signals that distinguish a real trend from a fad, and the timing windows that decide whether a trend launch pays back.
Spot trend curves in Amazon's own search data.
AMALYZE tracks monthly search-volume movement on every keyword. The acceleration is visible before the SERP saturates — early enough to launch, with data instead of intuition.