Return & Satisfaction Policy
Our commitment to your satisfaction. This policy outlines what you can expect when you work with GigMill to build AI agents for your business.
✅ Satisfaction Guarantee
Every GigMill AI agent delivery includes the following to ensure your satisfaction:
What's Included in Every Delivery:
- 2 rounds of revisions -- Adjustments to align the agent with your specifications
- 7-day testing period -- Time to try the agent in your workflow
- Documentation -- Setup guides and usage instructions
- Basic training -- Walkthrough of agent capabilities and limits
"Satisfaction" means the deliverable functions as described in the project brief and meets the agreed-upon specifications. It does not mean we'll rebuild the entire project if your requirements change mid-project.
🛠 What Qualifies for a Fix or Redo
The following situations qualify for corrective action at no additional cost:
- Agent doesn't perform as described -- The AI agent fails to deliver the functionality outlined in your project brief
- Technical bugs or failures -- Critical errors that prevent the agent from functioning
- Missing features -- Functionality that was explicitly agreed upon but not delivered
- Quality issues -- Output that consistently falls below professional standards
Process: Contact us within 14 days of delivery with specific issues. We'll diagnose and fix within 5 business days.
❌ What Does NOT Qualify
The following are considered scope changes, not defects:
- "I changed my mind" -- Requests to fundamentally change what was originally agreed upon
- Scope expansion -- Adding features or capabilities beyond the original brief
- Client infrastructure issues -- Performance problems caused by your existing systems, third-party APIs, or hosting environment
- Post-delivery modifications -- Changes requested after the testing period has ended
- Use cases not in scope -- Running the agent for purposes not discussed in the original project
Note: Scope changes can be quoted as new projects. We'll provide a fair estimate before starting any additional work.
💰 Refund Policy
Our refund policy is designed to be fair while recognizing the custom nature of AI agent development:
| Scenario |
Refund |
| No work has started |
100% |
| Work started but nothing delivered |
75% |
| Delivered but fundamentally flawed |
50% |
| Delivered as specified, but you want something different |
0% |
| Requested revisions beyond 2 rounds |
0% |
Fundamentally flawed means the deliverable does not match the project brief in material ways -- not minor adjustments or preferences.
Refund requests must be submitted within 14 days of delivery. After 14 days, no refunds are available.
⚖ Dispute Resolution
Step 1: Contact Us First
If you're not satisfied, reach out directly. Most issues resolve with a conversation. We'll work to make things right.
Step 2: Mediation
If unresolved after 14 days, we agree to mediation before any other action. Both parties select a neutral mediator, and we'll split costs equally.
Step 3: Arbitration
If mediation doesn't resolve the dispute, we agree to binding arbitration under American Arbitration Association rules. This is faster and less expensive than court.
Timeline: Issues must be raised within 14 days of delivery. After 14 days, disputes cannot be raised, and you forfeit any refund claim.
🤖 AI-Specific Disclaimers
AI agents are different from traditional software. Please note:
- Periodic tuning may be needed -- AI agents may require adjustments over time as your business needs evolve or as AI models are updated
- Performance depends on training data -- The quality of outputs depends on the information you provide. Garbage in, garbage out
- No guarantee of specific outcomes -- We build agents that can achieve your goals, but we cannot guarantee specific business outcomes like revenue increases or conversion rates
- AI limitations -- AI agents may occasionally produce unexpected outputs. We'll work to minimize this, but some edge cases are inherent to the technology
- Third-party dependencies -- Many AI agents integrate with external services (APIs, databases). We can't control those services' availability or changes
Our commitment: We build AI agents using industry best practices and quality assurance. We stand behind our work. But AI is probabilistic, not deterministic -- results vary based on inputs, configuration, and use case.