Use paid queries as evidence about buyer language
Search-term data can show how people described a need before interacting with your ads. It can reveal useful questions, service terminology and irrelevant intent. It is a starting point for SEO research, not a direct map of every organic opportunity.
Work from a dated export with the account, market and campaign scope recorded. Remove personal information and any queries unsuitable for wider sharing. Preserve the original query wording in your working file so interpretation remains traceable.
The search terms report omits some terms with low query activity to protect privacy. Treat it as a partial view: a missing query does not establish that no demand exists.
Separate the query from the outcome
Record what is actually known about each query: impressions, clicks, cost and the conversion definition used. Where reliable qualification or sales information exists, add that context. A platform conversion count may represent a form action rather than a qualified opportunity.
Do not label a query a winner because it generated one incomplete or unverified conversion. Equally, a query with no observed conversion may have too little data to support a conclusion.
Map the task to an existing page first
Ask what the searcher wanted to accomplish. A service request may belong on a commercial service page. A comparison question may need an explanation. A practical calculation may fit an existing tool.
Inspect the page body before deciding coverage is missing. Record the existing owner and what specifically needs improvement. A new phrase does not automatically justify a new article, and a new format does not create a new search task.
Validate the organic opportunity
Look at current organic results for the intended market and inspect the page types that answer the task. Add current query-by-page evidence from Search Console when available. Search volume, organic competition and an ad click’s cost are different measures.
Use a separate evidence status for ideas that still need research. The worksheet offers improve existing, research new page, exclude and hold decisions. Keep the reason beside the decision so the next person can understand it.
A fictional decision example
A service company sees several paid queries asking whether a particular service includes inspection. The existing service page already describes the inspection but makes it difficult to find. The first action may be to clarify that section and measure the result, rather than launch an article competing for the same service request.
Another query asks how to compare two approaches before buying. If no current page answers that comparison, it may deserve a distinct guide after a fresh result-page review. Use the measurement framework to keep the evidence and business outcome clear.
Optional: use AI to suggest groups
Start with an export approved for the tool you plan to use, with personal or sensitive information removed. Ask for draft groups and the reason for each assignment. A reviewer should compare every proposed group with the original queries, flag ambiguous cases and check the actual pages before adding an action to the work list.
Keep the model’s labels separate from observed conversion and qualification data. Do not let a generated group create new pages, negatives or campaign changes automatically.
Take the worksheet with you
Use the blank worksheet for your own review. Any completed examples are clearly labeled as fictional.