SEO & GEO Researcher

Audits a domain's organic and AI search position for any topic — surfacing content gaps, ranking opportunities, and ChatGPT / AI Overview visibility versus competitors. The full-powered SEO + GEO research tool: built for deep, multi-faceted analysis with a data-rich strategic narrative that tells the topic's SEO story — not a light spot-check. Use it to decide where a domain should invest content effort to win both Google rankings and AI-era discovery. Works on any industry and any DataForSEO-supported market.

Core Capabilities

• Two input modes: (1) domain + topic-keyword patterns, (2) URL list upload (CSV or Excel, up to 100 URLs per run)
• Multi-domain comparison against competitor domains (auto-discovered if not provided)
• Keyword clustering by topic with per-cluster performance metrics
• Traffic estimation via CTR-by-position modeling
• Configurable noise filtering (login pages, social media, legal pages, etc.)
• GEO/AI visibility tracking: brand mention counts in LLM responses, top citing sources, AI Overview presence
• Live ChatGPT response scraping for spot-checks of specific queries

Good Query Examples

• "Compare tk.de rankings for stress and sleep keywords against aok.de and barmer.de in Germany" → Domain + patterns mode: clear domain, topics, competitors, and market — proceeds directly
• "Here are 80 URLs from Website Content Clustering about stress on tk.de — analyze their keyword rankings against aok.de and barmer.de" → URL list mode: precise page-level analysis across many pages, no substring guessing
• "Full SEO audit for personaldienstleister.de with focus on zeitarbeit and personalvermittlung keywords" → Domain with topic patterns, auto-discovers competitors
• "I have a CSV with 80 product page URLs from otto.de — run full keyword analysis against amazon.de and zalando.de" → URL list mode from CSV input
• "Check if bmw.de appears in ChatGPT responses for 'best luxury electric vehicles'" → GEO/AI visibility analysis with ChatGPT scraping
• "How visible is example.com in AI Overviews for cloud computing queries?" → GEO-focused analysis using LLM mention tools

Bad Query Examples

• "Check how otto.de ranks for fashion topics" → "Fashion topics" is too broad for topic patterns — the tool uses substring matching, not semantic search. Fix: provide 3-6 specific keyword stems like "sneaker", "kleid", "jacke", "sportschuhe", "wintermantel"
• "Compare bmw.com against all competitors" → "All" competitors isn't feasible — need specific domains. Fix: name 2-4 competitor domains explicitly, e.g., "mercedes-benz.com and audi.com"
• "Analyze tk.de, aok.de, and barmer.de for stress keywords" → Which domain is the target and which are competitors? Fix: "Analyze tk.de (our domain) against aok.de and barmer.de for stress keywords"
• "Run SEO analysis for docmorris.de on pharmacy keywords in English" → Location/language mismatch — docmorris.de is a German site, analyzing in English would miss most rankings. Fix: specify German market (location="Germany", language="de")

Use-Case Ideas

• Topic-level competitive intelligence — who owns which keyword clusters and where the domain has gaps
• AI visibility audit — domain presence in ChatGPT and AI Overviews versus competitors
• Content strategy planning — cluster-level gaps and opportunities drive content priorities
• Traditional vs. AI search gap — keywords where the domain ranks well in Google but is absent from AI responses

Limitations

• Domain + patterns mode uses substring matching, not semantic search — provide 3–6 specific keyword stems
• Both modes only surface pages that already rank for a relevant keyword (no discovery of un-ranked pages)
• URL list mode has no URL discovery — URLs must come from the user (CSV or Excel upload)
• Target 200–500 keywords per run; fewer than 50 means input is too narrow, more than 2000 means too broad
• One topic cluster per run — split runs to compare across different clusters

Input/Output Spec

Inputs (form fields):

• Target Domain (required): the primary domain to analyze, without https:// or www. (e.g., tk.de).
• Topic Patterns (optional): 3–6 comma-separated keyword stems for substring filtering (e.g., stress, schlaf, burnout, entspannung). Required for domain + patterns mode.
• Competitor Domains (optional): one per line. If left empty, competitors are auto-discovered via DataForSEO SERP analysis.
• URL list file (optional): CSV or Excel (.xlsx) with one URL per row. Up to 100 URLs per run — lists above the cap are silently truncated, so split larger lists across multiple runs. URLs must include https:// or http://. Provided, this triggers URL list mode and the target domain is auto-extracted.
• Country (required): DataForSEO location (e.g., Germany, United States, United Kingdom, France).
• Language (required): the language to analyze in (e.g., German, English, French). Internally mapped to a DataForSEO language code.
• Report Language (required): the language to write the final report in — independently selectable from the analysis language (e.g., analyze German rankings, deliver the report in English).
• Additional Instructions (optional): free-text steer (e.g., "focus on AI/GEO visibility only", "compare content strategies").

Outputs:

• Excel workbook (.xlsx) — keyword tables, cluster analysis, opportunities, methodology.
• Strategic markdown report — headline insight, competitive narrative, recommendations organized into three effort tiers (Quick Wins, Strategic Improvements, Long-Term Bets), and a GEO/AI visibility section when AI analysis is requested.

Recommended Workflow

For comprehensive analysis, run Website Content Clustering first to discover the full content landscape, then feed the output into SEO & GEO Researcher.

Chain A: WCC → URL list mode (recommended for large or unfamiliar topics)

1. Website Content Clustering discovers pages across the topic and returns a full URL list.
2. SEO & GEO Researcher (URL list mode) takes up to 100 of those URLs — fetches every keyword each URL ranks for, no substring guessing, no semantic gaps. If WCC returns more than 100 URLs, prioritize the most relevant 100 or split across runs.

Chain B: WCC → domain + patterns mode (for targeted analysis)

1. Website Content Clustering surfaces the thematic clusters.
2. Convert the cluster themes into topic patterns (e.g., ["stress", "burnout", "resilienz", "achtsamkeit", "schlaf", "meditation"]) and run SEO & GEO Researcher in domain + patterns mode. Broader than URL mode (catches domain-wide rankings) but subject to substring matching.

Skip WCC when: the user already provides a URL list (CSV / Excel / CMS export) or precise keyword patterns. WCC is most valuable when the topic is broad, the domain is unfamiliar, or the user says something general like "analyze health content."

Disambiguation Signals

• SEO & GEO Researcher vs. Page Ranking Insights — both analyze keyword rankings via DataForSEO, but serve different scales:
• SEO & GEO Researcher = large-scale, up to 100 URLs per run, keyword clustering, Excel output, traffic estimation, GEO visibility. Two input modes.
• Page Ranking Insights = up to 20 URLs, narrative markdown report, cannibalization detection, quick wins.
• Route to Page Ranking Insights when: the user has ≤20 URLs, wants a narrative content strategy report, needs cannibalization detection, or wants a quick analysis without Excel overhead.
• Route to SEO & GEO Researcher when: the user has 21–100 URLs or a CSV/Excel URL list, wants domain-wide topic analysis, needs Excel output with keyword clustering, wants multi-domain comparison, or asks about AI/ChatGPT visibility.
• SEO & GEO Researcher vs. GEO Sniffer — both deal with AI visibility, but from opposite directions:
• SEO & GEO Researcher = "Does my domain appear in AI responses?" — domain-first, quantitative measurement (mention counts, citation frequency, competitive rankings). Also scrapes actual ChatGPT responses.
• GEO Sniffer = "How does the AI interpret this topic?" — topic-first, qualitative discovery (search strings, trusted sources, brand recommendations with labels).
• Route to GEO Sniffer when: the user wants to understand how AI frames a topic, which brands it recommends, or what sources it trusts — without starting from a specific domain.
• Route to Advanced Web Research when the user wants broad multi-source web research, not SEO/GEO ranking data.
• Route to Statista when the user wants market statistics or industry forecasts, not search performance data.
• Route to News Research when the user wants recent news coverage, not search ranking analysis.