Skip to content
Nuvexa AI
Menu
Research · PRACTICAL TUTORIAL

Separate Etsy keyword ideas from measured demand

Build a phrase shortlist from product facts and keep measured evidence separate from AI suggestions.

Published by Nuvexa AI · Published · 3 min read

Start here

Build a phrase shortlist from product facts and keep measured evidence separate from AI suggestions.

  • Your actual product format and buyer use
  • Candidate phrases and the fact supporting each
  • Any genuine dated observations from your own seller records

Define the product before the phrase

A buyer searching for a pattern PDF may want instructions rather than a finished object. State format, relevant materials, size conditions and use before brainstorming. A phrase that changes those facts is unsuitable even if it sounds popular. The source of the product fact should remain available for review.

Classify the evidence

Keep generated ideas, visible public observations and actual dashboard measurements in separate groups. Nuvexa does not connect to your Etsy shop or establish live search volume from a shop name. A useful phrase suggestion can still be unmeasured. Label it accordingly rather than assigning an invented popularity score.

Match every recommendation with support

For each phrase, write the fact that makes it relevant. If there is no supporting fact, remove it or investigate the missing detail. A candidate implying a physical product, a licensed character or a skill level you have not checked changes the listing’s claim. Avoid treating competitor wording as proof of private sales.

Keep dates and limitations with metrics

When you have actual visits or orders, record their period, source and context. Changes in price, advertising and availability can affect a comparison. A before-and-after total alone does not prove a title caused more sales. Keep uncertain causal claims out of the conclusion.

Choose a useful next observation

The next step may be a clearer listing, a reviewed title or tracking the buyer questions that remain. Define what observation would change your decision. This method produces a shortlist and a review plan; it does not promise placement, demand or revenue.

Keep observations separate from keyword suggestions

Use a simple research record with four fields: the phrase, where you saw it, the date and what it describes about your product. Add a fifth field for questions that remain unanswered. A phrase appearing in a listing establishes that the seller used it; it does not establish search volume or buyer demand. If a tool supplies a metric, note that tool’s definition and scope. This keeps your final shortlist traceable without turning a suggestion into a marketplace statistic.

Worked example and reusable input

Use the stated inputs to reproduce checked local outputs. Replace sample facts when applying a brief to your own work.

Product facts: [verified format, contents and use]
Candidate phrase: [idea]
Supporting fact: [actual fact]
Measured demand: [source and date, or unknown]
Rejected implication: [anything the product does not provide]

Key takeaways

  • Ideas and measurements are different evidence.
  • Tie every phrase to a product fact.
  • Leave unknown volume or earnings unknown.