How to use the Search Term Report Classifier.
Classify the user's underlying need, then state the evidence. Keep uncertain terms available for review instead of forcing every query into a yes-or-no label.
Make the workflow fit your task.
Classify each term against the offer and record a short reason. Keep service, research, employment and other relevant intents separate, preserving uncertain terms so a reviewer can inspect them rather than lose them in a binary filter.
- What you provide
- Exported paid-search queries and business offer.
- What you get
- Relevant, irrelevant and review-needed query groups.
See the input and the result.
Illustrative input and output · a teaching example, not a live WebAct run
Example input
Offer: paid bicycle repair. Terms: bike repair near me; bike mechanic salary.
Completed example
Relevant service intent: bike repair near me. Irrelevant employment intent: bike mechanic salary.
Load this input into the prompt, then copy it to WebAct to try the task. Your result may differ from the illustration.
Decisions and troubleshooting.
Can one search term have both research and purchase intent?
Yes. Use an ambiguous or mixed label when the wording does not establish the next action. Additional evidence may resolve it.
Why is the classifier treating related words as relevant buyers?
Compare the underlying need with the actual offer. A mechanic salary query mentions the trade but seeks employment information.
Try it with your own source.
Replace the example with your material in the task prompt. Keep the requirements you need, then copy the task into WebAct.
Customize and copy the task ↑