Skip to main content

Frequently Asked Questions

Quick answers to common questions about AIQ.

General Questions​

What is AIQ?​

AIQ is an Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) analytics platform. It helps you understand how your brand is represented in AI-generated answers from ChatGPT, Claude, Gemini, and other LLMs.

Do I need coding experience to use AIQ?​

No! AIQ provides a user-friendly dashboard for:

  • Executing queries across multiple LLMs
  • Analyzing citation patterns
  • Tracking brand mentions
  • Generating reports

However, developers can extend functionality through the API and custom integrations.

How much does it cost to run?​

Application: Free to self-host (open source)

LLM API Costs (approximate per 100 queries):

  • Gemini: Free (60 requests/minute free tier)
  • Perplexity: $0.10
  • Claude: $0.30
  • GPT-4: $1.00

Start with Gemini's free tier for testing, then add other providers as needed.

Can I use AIQ in production?​

Yes! AIQ is production-ready and includes:

  • Multi-organization support
  • Role-based access control
  • Encrypted API key storage
  • Cost tracking and budgets
  • Scheduled reporting

Setup & Installation​

What are the system requirements?​

  • Node.js: 18+ (20+ recommended)
  • Database: Supabase (PostgreSQL)
  • Hosting: Vercel, AWS, or any Node.js host
  • Browser: Modern browser (Chrome, Firefox, Safari, Edge)

How long does setup take?​

  • Quick Start: 15 minutes (clone, configure, deploy)
  • Full Setup: 1-2 hours (including API keys, customization, team setup)

Do I need my own Supabase account?​

Yes. AIQ uses Supabase for:

  • PostgreSQL database
  • Real-time subscriptions
  • Row-level security
  • User authentication (future)

The free tier is sufficient for most use cases (500MB database, 50,000 monthly active users).

Can I use a different database?​

Currently, AIQ is built specifically for Supabase/PostgreSQL. Support for other databases would require significant refactoring.

Features & Capabilities​

Which LLM providers are supported?​

  • ✅ Anthropic Claude (Sonnet, Haiku, Opus)
  • ✅ OpenAI (GPT-4, GPT-4 Turbo, o1)
  • ✅ Google Gemini (1.5 Pro, 2.0 Flash)
  • ✅ Perplexity (Sonar, Sonar Pro)

More providers can be added by extending the client architecture.

Can I compare responses across providers?​

Yes! Execute the same query across multiple providers simultaneously and compare:

  • Citation differences
  • Brand mention patterns
  • Response quality
  • Latency and cost

Does AIQ track costs automatically?​

Yes. Every query execution tracks:

  • Input/output tokens
  • Cost per provider
  • Cumulative spending
  • Cost breakdowns by time period

View costs in the Analytics dashboard.

Can I schedule automated queries?​

Not yet in the current version. Scheduled query execution is planned for a future release. For now, you can:

  1. Use batch execution for multiple queries
  2. Save query sets for repeated use
  3. Implement custom scheduling via API (developer guide)

Data & Privacy​

Where is my data stored?​

  • Query history: Your Supabase database
  • API keys: Encrypted in your Supabase database
  • LLM responses: Your Supabase database

No data is sent to third-party services except the LLM providers you explicitly configure.

Are API keys secure?​

Yes. API keys are:

  • Encrypted at rest (AES-256-GCM)
  • Scoped per organization
  • Never exposed in client-side code
  • Stored only in your Supabase database

Can I delete my data?​

Yes. All data is in your Supabase database. You can:

  • Delete individual queries
  • Clear analytics data
  • Drop entire tables
  • Delete the entire Supabase project

Is data shared between organizations?​

No. Multi-tenancy is enforced via:

  • Row-level security (RLS)
  • Organization-scoped queries
  • Isolated API keys per org

Performance & Scaling​

How many queries can I run?​

Limited only by:

  1. Provider rate limits (e.g., Anthropic Tier 1 = 50/min)
  2. Supabase database limits (free tier = 500MB storage)
  3. Your budget (LLM API costs)

How long do queries take?​

Typical query latency:

  • Fast models (Haiku, GPT-4o-mini): 2-5 seconds
  • Standard models (Sonnet, GPT-4): 5-10 seconds
  • Reasoning models (o1, Claude Opus): 15-30 seconds

Batch execution runs queries in parallel for faster completion.

Can I run 1000+ queries per day?​

Yes, but consider:

  1. Rate limits: Spread queries over time or upgrade API tiers
  2. Costs: 1000 Claude queries ≈ $3-5
  3. Database: Ensure adequate Supabase storage
  4. Processing: Batch execution helps manage large volumes

Customization​

Can I customize the UI?​

Yes! The UI is built with:

  • Tailwind CSS: Easy styling customization
  • Shadcn UI: Component library
  • Next.js: Full control over pages/routing

Modify components in /components and pages in /app.

Can I add custom analytics?​

Yes. Create custom SQL queries or views in Supabase, then display results in custom React components.

Example: Top cited domains

CREATE VIEW top_cited_domains AS
SELECT domain, COUNT(*) as citation_count
FROM citations
GROUP BY domain
ORDER BY citation_count DESC;

Can I integrate with other tools?​

AIQ includes API routes that can be called from:

  • Zapier/Make (webhooks)
  • Custom scripts (API calls)
  • Other applications (REST API)

See Developer Guide for API documentation.

Troubleshooting​

Queries aren't executing - what should I check?​

  1. API Keys: Verify keys are configured in Settings
  2. Network: Check internet connection
  3. Provider Status: Visit provider status pages
  4. Rate Limits: Check if you've hit limits
  5. Console Errors: Open browser console (F12)

See Common Issues for detailed troubleshooting.

Analytics show 0 queries but I executed some​

  1. Hard refresh the page (Cmd/Ctrl + Shift + R)
  2. Check if queries actually succeeded (look for errors)
  3. Verify database RLS policies allow reads
  4. Check Supabase dashboard for actual data

Citations aren't being extracted​

Some providers don't always include citations. Try:

  • Perplexity: Best for citations (optimized for search)
  • Claude: Good citation quality with proper prompting
  • GPT-4: Varies by model version

Adjust prompts to request explicit citations.

Billing & Costs​

How do I control costs?​

  1. Set provider spending limits (OpenAI, Anthropic dashboards)
  2. Use free tiers for testing (Gemini)
  3. Choose cheaper models (Haiku vs Opus, o1-mini vs o1)
  4. Monitor usage (AIQ cost tracking dashboard)

What if I exceed my LLM provider budget?​

Providers will:

  1. Send budget alert emails (if configured)
  2. Throttle or stop API access
  3. Require payment/plan upgrade

Set alerts at 50% and 80% of your budget.

Are there any hidden costs?​

No application-level costs, but consider:

  • Supabase: Free tier sufficient for most; $25/mo Pro for heavy use
  • Vercel: Free for personal; $20/mo Pro for commercial
  • LLM APIs: Pay-per-use based on queries executed

Support & Community​

How do I report bugs?​

  1. Check Common Issues first
  2. Search GitHub Issues
  3. Open a new issue with:
    • Clear description
    • Steps to reproduce
    • Screenshots/error messages
    • Environment details

Can I contribute to AIQ?​

Yes! AIQ is open source. Contributions welcome:

  • Bug fixes
  • New features
  • Documentation improvements
  • Provider integrations
  • UI/UX enhancements

See CONTRIBUTING.md for guidelines.

Where can I get help?​

  1. Documentation: Comprehensive guides in /docs
  2. GitHub Discussions: Community Q&A
  3. GitHub Issues: Bug reports and feature requests

Roadmap & Future Features​

What features are planned?​

Upcoming features include:

  • Scheduled queries: Automated query execution
  • Advanced analytics: ML-powered insights
  • Content gap analysis: Identify missing citations
  • Competitive intelligence: Track competitor mentions
  • Team collaboration: Comments, sharing, annotations
  • API webhooks: Real-time notifications

Can I request features?​

Yes! Open a feature request on GitHub Issues with:

  • Use case description
  • Expected behavior
  • Why it's valuable
  • Any implementation ideas