Frequently Asked Questions

VisibilityRadar measures how often and how positively AI models (Claude, GPT-4o, Gemini, Perplexity, Grok, DeepSeek) mention your brand. You get an overall AI Visibility Score (0–100), per-model breakdowns, sentiment analysis, competitor comparison, and a model-specific strategy playbook.

AI models are increasingly how people discover brands, products, and services. When someone asks ChatGPT or Claude for a recommendation, brands with higher AI visibility get mentioned — brands with low visibility are invisible. GEO (Generative Engine Optimization) is the new SEO.

Every prompt in your analysis is sent to each active AI model. Per prompt, your brand earns 10 points if it is mentioned at all, plus 5 more if it is mentioned at least as often as every competitor — so the maximum is 15 points per prompt with competitors, 10 without. A model's score is the points earned divided by the maximum possible, out of 100, and your overall score is the plain average across models. Sentiment is reported separately and does not affect the score. The full formula, a worked example and the known limitations are published at /methodology.

We recommend monthly analyses to track trends. AI models update their training data periodically, so your score can change over time. Pro and Agency plans show month-over-month score changes automatically.

The Free plan includes 1 analysis with 2 AI models (Claude and GPT-4o), basic score breakdown, and limited recommendations. It's great for a quick first look at your brand's AI visibility.

Pro gives you 10 analyses per month across all 6 AI models, competitor comparison (up to 3 brands), full sentiment analysis, AI Strategy Playbook, AI Content Studio (5 blog posts/month), PDF export, weekly email digest, and MCP API access.

Agency includes everything in Pro plus unlimited analyses, up to 5 competitors per analysis, custom prompt library (50 prompts), white-label PDF reports, priority support, and MCP API access with higher rate limits.

Yes. You can cancel your subscription anytime from your Account Settings page. Your plan stays active until the end of the billing period — no partial refunds, no hidden fees.

No. Monthly analysis credits reset on the 1st of each month. We recommend scheduling regular analyses to get the most out of your plan.

We test Claude (Anthropic), GPT-4o (OpenAI), Gemini (Google), Perplexity, Grok (xAI), and DeepSeek. Free accounts get Claude and GPT-4o. Pro and Agency accounts get all 6.

The Playbook is a model-specific action plan generated by AI. For each model that gave your brand a low score, you get specific tactics — like 'build a Wikipedia presence' for Claude or 'earn reviews on G2' for Perplexity. Each action is prioritized by impact.

Content Studio generates 5 SEO and GEO-optimized blog posts tailored to your brand and the prompts AI models use to discover it. The posts are designed to improve your brand's citation chances in AI responses over time.

Site AI Optimization runs a 10-point audit of your website — checking schema markup, brand density, meta tags, and other signals that AI models use to understand and trust your brand. Available on Pro and Agency plans.

Some AI models (especially Gemini) have safety filters that avoid making direct brand comparisons. If a model declines to respond to brand-specific prompts, it may return a 0 score. This is a known limitation and we're continuously improving prompt strategies to minimize it.

Yes. Pro plans support up to 3 competitors per analysis, Agency plans support up to 5. You'll see side-by-side AI Visibility Scores, per-model breakdowns, and alerts when a competitor pulls ahead.

MCP (Model Context Protocol) is an open standard that lets AI assistants like Claude Desktop, Cursor, and Windsurf use external tools. With the VisibilityRadar MCP server installed, you can analyze brands, check history, and compare competitors directly from your AI chat — no browser needed.

1. Go to Account Settings and generate an API key under 'MCP API Keys'. 2. Install the package with: npx visibilityradar-mcp. 3. Add the config to your Claude Desktop or Cursor settings file with your API key. Full instructions are shown in Account Settings.

Yes. Every analyze_brand call via MCP counts as one analysis and deducts from your monthly quota — exactly the same as running an analysis from the dashboard. Results are also saved to your dashboard automatically.

Pro: up to 5 analyses per day via MCP. Agency: up to 20 analyses per day via MCP. Monthly limits still apply.

No. MCP API access requires a Pro or Agency plan. You can upgrade at any time from the Pricing page.

Yes, your analysis history is stored securely in our database so you can track trends over time and access past results from your dashboard. You own your data and can delete your account at any time.

API keys are shown only once at creation and stored as a hash in our database. If you lose your key, you'll need to revoke it and create a new one. Never share your API key publicly.

We send your brand name, market, and competitor names to AI models to generate responses — this is the core of the analysis. We do not send any personal user data to AI models. Our Privacy Policy covers this in detail.