Vizzybl LogoVizzybl LogoVizzybl

Brand Protection In Depth

Explore all Brand Protection tabs: Content Index, Adversarial, Fact Check, Red Team, Simulated RAG, Gap Analysis, and Internal Links.

Last updated

Brand Protection in depth

Brand Protection monitors how AI platforms represent your brand and defends against misinformation, adversarial attacks, and content gaps. This page covers each analysis tab in detail.

For initial setup, see Brand Protection.

Indexing your site

Before running any analysis, index your website so Vizzybl has content to work with.

  1. Click Index Site in the Brand Protection header
  2. Enter your domain
  3. Choose a crawl depth — Single page, Top 20, Top 50, Top 100, or Full site. Top 100 is the default on paid plans; free tenants are capped at Top 20.
  4. Click Start Indexing

A progress indicator shows the crawl status in real time. Once complete, your indexed pages appear in the Content Index tab.

Content Index tab

Shows all indexed pages across your domain and competitor domains.

For each indexed domain you see:

  • Domain name with page count
  • Last indexed date
  • Index status (complete, in progress, or failed)

Click into any domain to see individual pages with their titles and indexing metadata. Re-index at any time to capture content changes — monthly re-indexing is recommended.

Adversarial tab

Detects security flags and potential adversarial content in AI responses about your brand.

What it analyzes

Vizzybl scans AI responses for patterns that could indicate:

  • Prompt injection attempts — Malicious content trying to manipulate AI behavior
  • Harmful content — Responses containing dangerous, hateful, or explicit material
  • Misinformation — False claims about your brand
  • Data leakage — Sensitive information exposed in responses

Reading the results

Each flag includes a severity level, category, and the specific response that triggered it. Use these flags to identify patterns in how AI platforms might misrepresent your brand.

Fact Check tab

Verifies the factual accuracy of claims AI platforms make about your brand.

Vizzybl compares AI-generated statements against your indexed content to identify:

  • Confirmed facts — Claims that match your published content
  • Contradicted claims — Statements that conflict with your content
  • Unverifiable claims — Statements that cannot be confirmed from your indexed pages
  • Hallucinated information — Claims with no basis in any known source

Focus on contradicted claims first — these represent active misinformation that could mislead your audience.

Each claim carries how widespread it is (the share of checked answers repeating it), which engines said it, and the ground-truth statement it contradicts.

Claim Breakdown

For every inaccurate claim, Fact Check traces the web page most likely feeding it. Citation Share is the proportion of the answers carrying a claim that also cite a given page — a page cited by every one of them is far more likely to be the cause than a page cited by one.

Each page is categorised by who controls it, which determines the remedy:

CategoryMeaningSuggested action
OwnedYour own siteUpdate content — correct the page directly
EarnedA review site, forum or publicationOutreach — ask the publisher to correct it
CompetitionA competitor's pageCreate content — publish the correct answer yourself
UncitedThe answers cited nothing at allCreate content — the model is drawing on training data

Claims we are not confident enough about are marked Monitor rather than given an action — acting on a shaky verdict is worse than waiting.

Any row can be filed to your Actions board with one click. The card carries the false claim, the ground truth, how widespread it is and the offending page, so it can be worked without opening Vizzybl.

Red Team tab

Simulates adversarial attacks against your brand to identify vulnerabilities before bad actors exploit them.

Assessment types

TypeWhat it doesAvailability
StandardGenerates adversarial prompts and tests them against your indexed contentAll plans
LiveSends adversarial prompts to real AI engines (ChatGPT, Perplexity, Gemini) and analyzes their actual responsesUltra / Enterprise

Each assessment produces a Red Team Score, a per-category breakdown, and concrete recommendations for content you should strengthen.

Typical attack patterns tested

  • Competitor comparison prompts designed to position your brand negatively
  • Questions about known vulnerabilities or past incidents
  • Prompts that test for hallucinated negative information
  • Edge cases where AI platforms might confuse your brand with others

Simulated RAG tab

Tests how a Retrieval-Augmented Generation system would respond to queries using your indexed content as the knowledge base.

Why it matters

Many enterprise AI deployments use RAG to ground responses in specific content. This tab simulates that process so you can see:

  • Which of your pages would be retrieved for common queries
  • How well your content answers the query
  • Whether gaps in your content lead to incomplete or inaccurate answers

Enter test queries to see which pages your content surfaces and how a RAG system would synthesize them into a response.

Gap Analysis tab

Identifies content gaps by comparing your indexed content against a competitor's.

Running a gap analysis

  1. Enter a topic (e.g. "sustainable fashion")
  2. Choose how many subtopic queries to auto-generate (5, 10, 15 or 20)
  3. Click Run Analysis
  4. Select a competitor domain (must be indexed)
  5. Click Run Analysis

The results show:

  • Your Score, Competitor Score and Gap — three summary cards scoring overall coverage out of 100
  • Topic Breakdown — a table scoring you against the competitor per auto-generated subtopic, with a severity badge (critical: competitor ahead by >30%; moderate: 10–30%; minor: 1–10%; tied: within ±1%; ahead: you lead by >1%)
  • Recommendations — a prioritized list of actions with suggested content for each
  • Topics you cover that your competitor doesn't — Your content advantages
  • Shared topics — Areas where you compete directly

Use these gaps to prioritize content creation. Closing gaps on high-value topics improves your AI citation likelihood.

Analyzes the link structure within your indexed pages and suggests internal links to improve topical authority.

For each suggestion you see:

  • Source page — The page that should contain the link
  • Target page — The page being linked to
  • Suggested anchor text — AI-recommended link text

Stronger internal linking helps AI crawlers discover related content and reinforces your authority on connected topics.

  1. Index your site (Top 50 is a good starting point)
  2. Check Fact Check for contradicted or hallucinated claims
  3. Run a Gap Analysis against your top competitor
  4. Use Simulate RAG on your highest-priority queries to confirm coverage
  5. Run a Red Team assessment before a launch or campaign
  6. Re-index monthly or whenever you publish significant content changes

Next steps