AEO Visibility Monitor

Recurring visibility tracking across AI answer engines. Runs the aeo-visibility check on a schedule, compares results to the last run, and surfaces what changed — brand mentions gained or lost, competitor movements, new query opportunities.

Built on top of: aeo-visibility skill. This skill adds the recurring cadence, delta analysis, and trend reporting layer.

When to Use

• "Track our AEO visibility over time"
• "Are we appearing more or less in AI search?"
• "Set up monthly AEO monitoring for [client]"
• "What changed in our AI search visibility since last month?"
• "Are competitors gaining ground on AI answer engines?"

Phase 0: Intake

1. Company name and website URL
2. 2-3 main competitors to track visibility against
3. Run cadence: monthly (default), bi-weekly, or weekly?
4. First run or recurring run?
   • First run: Will profile the company and generate queries (takes 15-30 min)
   • Recurring run: Reuses existing profile and queries (faster)
5. Any new queries to add since last run? (Optional)

Phase 1: Check for Existing Profile

Look in clients/<client-name>/aeo-visibility-reports/:

clients/<client-name>/aeo-visibility-reports/
├── profile.json        # Company + competitor profile
├── queries.json        # ~100 test queries
├── history.csv         # Summary of all runs
└── runs/
    ├── 2026-02-01T08-00-00/
    └── 2026-03-01T08-00-00/

If profile exists: Skip to Phase 3 (run check). If not: Run Phase 2 (profile setup).

Phase 2: First-Time Setup (Profile)

Run aeo-visibility profile command:

node skills/aeo-visibility/src/cli.js profile <client-name> --url=<website_url>

This will:

• Research the company and identify competitors
• Generate ~100 diverse test queries covering:
  • Brand queries (direct mention)
  • Category queries (what you do)
  • Problem queries (the pain you solve)
  • Comparison queries (you vs competitors)
  • Use-case queries (specific scenarios)

Before running the full check: Review and edit queries.json to:

• Remove irrelevant queries
• Add high-priority queries specific to this client's positioning
• Ensure competitors are correctly identified

Phase 3: Run Visibility Check

node skills/aeo-visibility/src/cli.js run <client-name>

Runs all queries against all 4 engines. Creates a dated run folder with:

• report.md — full analysis
• results.csv — raw data per query per engine
• raw-responses.json — full API responses

Cost per run: ~$10-30 (100 queries × 4 engines × ~$0.025-0.075/query)

Phase 4: Delta Analysis (Core Value of This Skill)

After the run completes, compare the new results.csv against the previous run's results.csv.

Metrics to Compare

For each engine (ChatGPT, Perplexity, Gemini, Claude):

Metric	Definition
Mention rate	% of queries where brand is mentioned
Prominence rate	% of mentions that are "high" prominence
Avg sentiment	Average sentiment score across all mentions
Competitor share	% of queries where competitor is mentioned

Change Flags

Change	Threshold	Severity
Mention rate ↓	>5% drop	🔴 Alert
Competitor mention rate ↑	>5% increase	🟡 Watch
Prominence drop	High → Medium or Low	🟡 Watch
New queries with 0 mentions	Any new "missed opportunities"	🟡 Watch
Mention rate ↑	>5% increase	🟢 Win

New Opportunity Detection

Compare queries where:

• You: not mentioned AND competitor: mentioned → These are displacement queries to target
• You: mentioned, low prominence → Content optimization opportunity

Phase 5: Output Format

# AEO Visibility Report — [DATE]
Client: [Name] | Previous run: [DATE] | Change period: [N days]

---

## Summary

| Metric | Previous | Current | Change |
|--------|----------|---------|--------|
| Overall mention rate | [X%] | [X%] | ▲/▼ [X%] |
| High-prominence rate | [X%] | [X%] | ▲/▼ [X%] |
| Avg sentiment | [X] | [X] | ▲/▼ |
| [Competitor] mention rate | [X%] | [X%] | ▲/▼ [X%] |

**Net verdict:** [Improving / Stable / Declining]

---

## Changes Since Last Run

### 🟢 Wins
- [Engine]: Mention rate ↑ from [X%] to [X%]
- Now mentioned for: "[query example]" — was not before
- ...

### 🔴 Alerts
- [Engine]: Mention rate ↓ from [X%] to [X%]
- Lost visibility on: "[query example]"
- [Competitor] gained mentions on [N] queries where you dropped

### 🟡 Watch
- [Competitor] mention rate up [X%] on [Engine]
- [N] new queries added where you have zero mentions

---

## Engine Breakdown

### ChatGPT
- Mention rate: [X%] ([▲/▼ X%])
- Prominence: High [X%] / Medium [X%] / Low [X%]
- Top performing queries: [list]
- Missed queries (competitor mentioned, you not): [list]

### Perplexity
...

### Gemini
...

### Claude
...

---

## Competitor Movements

| Competitor | Mention rate change | Engines gaining | Notes |
|-----------|--------------------|--------------|----|
| [Name] | ▲ [X%] | ChatGPT, Perplexity | [context] |
...

---

## Top Opportunity Queries

Queries where a competitor appears but you don't — highest traffic potential:

1. "[Query]" — [Competitor] appears, you don't | Est. query volume: [range]
   Suggested fix: [content or PR action to get mentioned here]

2. "[Query]" — ...

---

## Recommended Actions

### This Month
1. [Specific action to recover lost visibility on [engine]]
2. [Content to publish targeting [opportunity query]]

### Next Quarter
1. [Structural PR / content strategy to improve overall mention rate]

Save to clients/<client-name>/aeo-visibility-reports/runs/[YYYY-MM-DD]/monitor-report.md.

Also update clients/<client-name>/aeo-visibility-reports/history.csv with the run summary row.

Scheduling

# Monthly on the 1st at 8am
0 8 1 * * node run_skill.js aeo-visibility-monitor --client <client-name>

Cost

Component	Cost per run
First-time profile	~$0.50
Full visibility check (100 queries × 4 engines)	$10-30
Delta analysis (LLM reasoning)	~$0.50-1
Total per monthly run	~$11-32

Environment Variables Required

OPENAI_API_KEY=...
ANTHROPIC_API_KEY=...
GOOGLE_API_KEY=...
PERPLEXITY_API_KEY=...

Tools Required

• Upstream skill: aeo-visibility
• API keys for all 4 engines (see above)

Trigger Phrases

• "Run the monthly AEO visibility check"
• "How has our AI search visibility changed?"
• "Are we gaining or losing ground in AI answer engines?"
• "Set up AEO monitoring for [client]"