Hire agents with personality
Infuse your brands style with persona driven agents
Configuring Your AI Agent with Chatform AI
Purpose
At Chatform AI, our mission is to deliver the highest quality player support. Too often, outsourced support agents or overwhelmed internal teams don’t have the time to be as patient and empathetic as players deserve. Not so with Chatform AI. Our default agent is designed to be friendly, empathetic, and helpful.
However, we also empower brands to customize their agents as if they were hiring a team of human agents. The goal is to create agents that sound neither impersonal nor robotic. Instead, your AI agent should feel like a warm, humble, polite, and empathetic team member—friendly, informative, helpful, and positive.
Best Practices for Configuring Your Agent
Attributes play a key role in shaping how the agent interacts with players and handles conversations. Some agents may need a professional tone, while others can be more playful or even flirty. Consider the type of experience you want your players to have and choose attributes accordingly.
Agent Instructions
This is where the magic happens. Think of agent instructions as sending out a company-wide memo to all your human agents with clear guidelines on how to handle tickets. These instructions dictate how the AI agent responds, ensuring alignment with your brand’s voice and policies.
Here are some examples to inspire you:
Example Use Cases
- Ongoing Promotions: Instruct the agent to mention current promotions when denying goodwill bonus requests.
- Consistent Sign-Offs: Direct the agent to sign off every message with a specific signature.
AI Response Customization Guidelines
| Guideline | How It Shapes Responses | Additional Notes |
|---|---|---|
| Adapt responses for regional preferences. | Ensures language, tone, and phrasing feel native to the target market. | Example: More expressive for Italian audiences, more reserved for Northern European ones. |
| Prioritize clarity over excessive politeness. | Keeps messages direct and helpful rather than overly formal. | Especially useful for troubleshooting and FAQs. |
| Highlight solutions before acknowledging issues. | Prevents responses from feeling negative or overly focused on problems. | Example: “You can reset your password by clicking here” instead of “Sorry for the trouble, you can reset…” |
| Use engaging prompts to keep conversations going. | Encourages users to share more details naturally. | Example: “Are you looking for a refund or an exchange?” instead of “How can I assist you?” |
| Avoid one-size-fits-all phrasing. | Keeps responses dynamic and varied to prevent repetitive messaging. | Helps AI sound more human and less scripted. |
| Inject subtle enthusiasm where appropriate. | Makes the AI feel more engaging, especially in customer interactions. | Example: “Great choice! Let me help you with that.” instead of “Sure, I can help.” |
| Offer proactive suggestions in responses. | Anticipates user needs rather than waiting for follow-up questions. | Example: “If you need help tracking your order, here’s how.” |
| Use inline links instead of long instructions. | Makes responses concise and easy to follow. | Example: “Click [here] to update your settings.” instead of “Go to Settings, then click Account, then select Preferences…” |
| Tailor responses based on user behavior history. | Provides personalized replies based on past interactions when possible. | Example: Recognizing repeat customers and acknowledging past issues. |
| Keep response length appropriate for the context. | Ensures messages aren’t too short (unhelpful) or too long (overwhelming). | Quick replies for chat, detailed responses for emails. |
| Acknowledge when AI doesn’t have an answer. | Avoids misleading users with incorrect or vague responses. | Example: “I don’t have that info, but I can connect you with a human agent.” |
| Make instructions actionable instead of vague. | Ensures users know exactly what to do next. | Example: “Go to ‘Account Settings’ and click ‘Billing’” instead of “Check your settings.” |
Crafting Effective Instructions for AI
While Large Language Models (LLMs) are proficient at following directions, they aren’t infallible. Instructions may sometimes be misinterpreted or inconsistently applied. To enhance reliability, use firm and explicit language—words like “Always” and “Never” help reinforce expected behavior.
By carefully fine-tuning your AI agent, you can deliver a player support experience that is both intuitive and tailored, ensuring every interaction reflects your brand’s unique voice and values.