Documentation
GEO Tracking
GEO stands for Generative Engine Optimization: making sure AI answer engines such as Claude, ChatGPT (GPT), and Gemini actually mention, cite, or recommend your brand when someone asks a relevant question. Traditional SEO earns you a ranking on a search results page; GEO earns you a place inside the AI-generated answer itself, which is increasingly where people stop looking any further.
The GEO section of your project runs a set of tracked prompts against those three models on a schedule, records whether your brand comes up in the answers, and turns the results into a citation score, a readiness score for how easy your site is for AI crawlers to read, and a prioritized roadmap of what to fix next.
Tracked prompt
Sent to Claude, GPT & Gemini
Citation detected
Score & roadmap
The GEO overview
The Overview tab is the default landing page for GEO and opens with an AI Overview panel: a written briefing, generated on request, that explains in plain language where you stand, why your citation score is what it is, and what to do next. It sits above the raw numbers on purpose, since the score alone doesn't tell a new user much.
Below it, a large citation-score card sits next to one card per model (Claude, GPT, and Gemini), each showing how many of that model's answers mentioned your brand out of the total probes run, plus your average position in the answer when you were mentioned (shown as a dash if a model never gave you a positioned mention in the selected window). You can switch the window between the last 7, 30, or 90 days.
A trend chart plots citation activity per model over that window, and a share of voice panel compares how often your brand is cited against the competitors you're tracking, with a link through to the full competitor breakdown.
The AI Overview also flags whether your tracked prompts themselves are well-formed (a score built on a single prompt that just asks "do you know Acme?" isn't measuring the same thing as one built on prompts a real customer would type) and links straight to the full roadmap for the detailed action list.
Tracked prompts & queries
A tracked prompt is a natural-language question you expect a customer might ask an AI assistant, for example, "what's the best project management tool for small agencies?" Each prompt runs against the models you choose for it (Claude, GPT, Gemini) on a set frequency, though the cadence that actually runs is the slower of what you requested and what your plan allows. Prompts are grouped by language rather than by country, since an AI model has no location. A German prompt scores the same whether the visitor is in Germany or Austria.
On a brand-new project, prompts are generated automatically from your first crawl once it finishes, so you'll see a "generating" state briefly rather than an empty list. Beyond that, you can add prompts yourself or pull from AI-generated suggestions, each with a short rationale for why it's worth tracking: promote a suggestion to start tracking it, or dismiss it if it's not relevant.
The page opens with at-a-glance stats: how many prompts are tracked and active, how many are currently cited (your brand appeared in the latest answer from at least one of its models), your overall mention rate, and when the last probe ran, plus filters for cited, uncited, and paused prompts, and a search box.
Expanding any prompt reveals its probe history: each run shows which model answered, whether your brand was mentioned, any competitors named instead, the answer's sentiment (positive, neutral, or negative), which URLs were cited (marked as yours or not), and an excerpt of the answer. Note that only the paragraphs carrying a link or a brand mention are kept, so a long answer may show as a partial excerpt rather than the full text.
GEO readiness score
The readiness score measures how easy your site is for AI crawlers and answer engines to read and trust, separate from whether they currently cite you. It's built from a set of health checks grouped into categories: crawler access (whether bots like ClaudeBot, OAI-SearchBot, ChatGPT-User, Google-Extended, and PerplexityBot are actually allowed to fetch your pages), discoverability, structured data, content signals, search inclusion (whether you're indexed by search engines that AI tools lean on, such as Bing and Brave), and content style. Each check comes back as pass, warn, or fail, with a plain-language explanation and a concrete fix.
An expandable best-practices panel underneath the checks gives copy-paste guidance: a sample robots.txt that explicitly allows the AI crawlers listed above, and a template for an llms.txt file that summarizes your site for AI tools the way a sitemap does for search engines. You can re-run the checks at any time with the recheck button.
A second, page-level card scores individual crawled pages for quotability: whether a page opens with a clear summary, uses tables and FAQ sections, is chunked under clear headings, carries schema markup, and reads as reasonably fresh. This analysis re-fetches pages live, so it only runs when you ask for it rather than automatically.
Recommendations
The Recommendations tab leads with a generated roadmap: an ordered, tickable list of actions written specifically for your site, each with a short explanation, a concrete first step, and ratings for impact, effort, and confidence in the signal behind it. Some actions you can finish yourself; others are flagged as depending on outreach to a third party (for example, getting listed somewhere). A posture badge (from absent to leading) summarizes where the roadmap judges you to stand overall.
You can check off items as you complete them or dismiss ones that don't apply. Both stay visible in collapsed sections afterward rather than disappearing, so progress is visible over time. A progress bar tracks how much of the actionable list is done, and you can export the whole roadmap as a CSV or regenerate it once your site or scores have changed.
A prompt audit section reviews the tracked prompts themselves (flagging any that are poorly worded with a suggested rewrite, and proposing additional prompts worth tracking), with a link back to Tracked prompts to act on them.
Below the roadmap, the page also lists the raw signals it was built from: any failing or warning readiness checks (with the same fixes shown on the Readiness tab, plus step-by-step submission guides for Wikipedia, Bing, and Brave where relevant), and any tracked prompts you're actively being probed on but never mentioned in.
