Local Business Intelligence
Convert location intelligence into useful creator, agency and business decisions.
This skill is a workflow layer, not a scraping engine.
Required References
Read:
references/EVIDENCE-POLICY.mdreferences/MAPS-AI-OPPORTUNITY-PLAYBOOK.md
For a new architecture or questionable data flow, use maps-policy-guard first.
Modes
Market Density / Expansion
Use aggregate or permitted live data to answer:
- How dense is this category in the target area?
- How does one area compare with another?
- Where does a service appear under-represented?
- Which hypotheses should be investigated further?
Prefer Places Aggregate API for density/count questions.
Do not present place density as guaranteed revenue or investment return.
Live Place Discovery
Use live place search for a narrow, user-requested task.
Examples:
- find suitable cafes for a creator shoot
- compare nearby venues
- shortlist businesses that match a campaign
- plan a route across selected places
Return only what is necessary for the task.
Do not construct a permanent bulk lead database from Google Maps content.
Owned Business Profile Performance
When the user has authorized access to managed locations, analyze:
- Maps / Search impressions
- website clicks
- calls
- direction requests
- supported search-keyword impressions
- review themes
- response gaps
- period-over-period changes
Separate observed metrics from interpretation.
Review Intelligence
For authorized / user-provided reviews:
- recurring praise
- recurring friction
- unanswered questions
- service confusion
- local language
- content ideas
- response-draft suggestions
Keep replies as drafts unless the user explicitly authorizes a supported write action.
Creator Location Scout
Inputs:
- city / area
- creator niche
- visual concept
- time window
- practical constraints
Output:
- short venue list
- why each place fits
- opening / access considerations when available
- route logic
- Reels / Shorts concept
- shot list
- collaboration angle
- fallback option
Local Sponsor-Fit Shortlist
Inputs:
- creator audience
- niche
- campaign idea
- geography
- brand exclusions
Output:
- small human-review shortlist
- fit rationale
- public evidence
- possible collaboration concept
- research gaps
Do not auto-send mass outreach.
Local Content Gap Engine
Combine first-party / authorized signals such as:
- GBP search-query data
- GBP performance metrics
- review themes
- business FAQs
- permitted local category context
Create:
- high-intent content gaps
- FAQ themes
- Reels concepts
- carousel concepts
- local landing-page briefs
- Pinterest / visual-search angles where relevant
- KPI plan
Evidence Labels
Every important claim should be labeled implicitly or explicitly as:
- Observed
- Inferred
- Recommended
Output
Local Business Intelligence Brief
Objective
- Business / creator:
- Geography:
- Mode:
- Time window:
Evidence
| Signal | Source | Date | Observation | Confidence |
|---|
Findings
Prioritize actionable findings.
Recommended Tests
| Test | Why | Effort | KPI |
|---|
Content / Campaign Opportunities
Only when relevant.
Data & Policy Check
- source:
- persistence:
- authorization:
- human approval required:
- policy status:
Next Decision
Choose the smallest useful next action.
Core Principle
Use local data to make a better decision, not to manufacture a larger database.