Choose which SEO opportunities to work on first

Use this guide to build a prioritized SEO work list from ranking opportunities, CTR gaps, declining pages, and underperforming queries. The assistant returns the evidence, estimate basis, and next review for each item.

When to use this use case

Use this use case when you need to choose work across several kinds of finding. The tool ranks click opportunities, but it does not know your business priorities or implementation effort.

What to ask

Replace the domain, dates, and business context before using this prompt.

Build an SEO work list for example.com using the 28 days ending September 30, 2026. Start with ranking and CTR opportunities. Add declining pages and visible content gaps if the broader call completes. Our priorities are [products, services, or audiences], and our capacity is [available effort]. Show the page or query, evidence, estimate assumptions, and next review for each item. Keep observed losses separate from projected gains. Do not add overlapping estimates or run paid topic discovery.

How the assistant runs it

The assistant starts with a smaller queue before adding more sources:

  1. Call get_server_status, list_sites, and get_connection_status to confirm the site and coverage.
  2. Call get_workspace_brand_terms with the resolved workspace ID. Check the terms used for branded-query exclusions.
  3. Call score_seo_opportunities with types set to low_hanging,ctr for the first review.
  4. Request low_hanging,ctr,decay,gap if you need the broader queue. Read sources to see which types ran, failed, or were skipped.
  5. Inspect selected findings with their named tools. Use get_query_pages for queries, get_page_queries with full URLs for pages, and detect_keyword_cannibalization before choosing competing-page work.
  6. Add business relevance and an effort assessment from supplied context or page review. Keep those judgments separate from the tool's click ranking.

Example tool input

This input requests a ranking and CTR queue without decay or gap analysis.

json
{  "tool": "score_seo_opportunities",  "arguments": {    "site": "example.com",    "to": "2026-09-30",    "windowDays": 28,    "types": "low_hanging,ctr",    "excludeBranded": true,    "minImpressions": 100,    "limit": 20,    "searchVertical": "web"  }}

What the result includes

The work list contains the following information:

  • Query or page, entity type, and relevant engine.
  • The signals that produced the finding.
  • Estimated clicks, estimate basis, and confidence.
  • Business relevance and effort information when available.
  • The next investigation or page review.
  • Sources evaluated, omitted sources, and data limits.

How to read the result

The tool uses the maximum of overlapping signals on the same entity, not their sum. A query and a page remain separate entities, so the queue does not remove every overlap between page and query opportunities. Do not sum the queue into a promised traffic gain.

Decay estimates describe measured losses. Ranking and CTR estimates project performance using the site's own CTR data. Check benchmark quality before treating a large projection as realistic. A null estimate means the tool could not price the item, not that it has no value.

The ranking reflects the visible candidates and sources that ran. It is not a complete list of every SEO action. Call the individual tools or reduce the sources if a broader request times out. Inspect pages and intent before treating multiple ranking pages as a reason to consolidate them.