This lesson builds the three competitor lists the module depends on, from data rather than assumption. It runs each one on the demo site, explains what each measures and shows how to reconcile them when — as they should — they disagree.
What this lesson covers
- Current-state competitors from shared ranking keywords
- Aspirational competitors from a fixed list of target queries
- The AI-citation competitor set
- Reading visibility and position separately
- Reconciling three lists
- Cost and repeatability
The current-state list comes from ranking-keyword overlap, sorted by shared keywords rather than any authority metric. On the demo site it was led by Stack Overflow, GeeksforGeeks and ResearchGate, because of the 148 US keywords the main site ranks for, only 21 contain an SEO term. That is a classification finding: the site is read as a data-science resource. The aspirational list starts from twelve queries the business wants to own and measures who is visible for them, and the lesson shows why visibility and average position must be read separately — a site can appear often and rank low, or rarely and rank high.
The third list comes from AI answers. On the demo site, seventeen AI mentions surfaced Surfer SEO and learningseo.io, two genuine competitors that appeared in neither ranking list. The lesson explains why citation and ranking behave differently, why the three lists are expected to disagree, and how to choose the comparison sites for the gap matrix from the overlap. The whole analysis cost nine cents of API credit, and every call is recorded so it can be repeated.
A written practical follows, running the same three lists through Claude and DataForSEO’s official MCP server with spending rules set before any call.
The lesson also covers how to read the results without overreaching. Ranking-keyword overlap describes how a search engine classifies a site today, not who the business considers a rival; visibility across a query set describes presence, not quality; and AI citations describe what assistants quote, not what they rank. Each list answers its own question, and the useful decisions come from comparing them rather than choosing one.
Location matters too: the demo data is US and English while the site is UK-based, and the lesson states that caveat on screen, as every competitor report should.
This lesson continues Module 4, Competitive Entity Research & Gap Analysis, in The AI-powered Entity Audit. It suits SEOs who need competitor research grounded in data and cheap enough to repeat every quarter.