Building a great support team takes time and effort. New team members need to learn about the business, how to interact with your customers, and how to approach resolving and escalating issues. It can take weeks to get a new team member trained and fully up to speed.

But what if the process could take minutes instead of weeks? Just like a new team member, Fin, Intercom’s AI Agent, can also be trained on how to interact with customers. But with Fin Guidance, this training happens instantly.

Fin Guidance is the new way to instantly customize Fin’s behavior to ensure every customer gets consistent, high-quality support. You train it the same way you onboard a new hire – by giving it clear instructions on how to talk to customers, clarify requests, and when to escalate. No technical setup. No complicated rules. Just guide Fin using natural language.

Customize Fin’s behavior in minutes

Every company supports customers differently. You have your own voice, your own processes, and your own internal policies. With Fin Guidance, you can train Fin in all of that – just like you would a new teammate.

You don’t need technical tools or scripts to do it. You simply write natural language instructions that shape how Fin responds. Here are some of the ways you can train Fin’s behavior using guidance.

Define Fin’s communication style

Make sure every reply sounds like it came from your team: on-brand, clear, and high-quality. Just like with a human agent, the more context you provide, the better Fin can represent your team. Here are real examples of how customers are guiding Fin’s language and tone:

• Use the right words: “Use clear, straightforward language and avoid jargon or buzzwords. For example, say ‘easy’ instead of ‘frictionless,’ ‘help’ instead of ‘enable,’ and ‘start’ instead of ‘onboard.’”
• Use the right language: “Always write in British English, using spelling like ‘colour’ instead of ‘color.’ Avoid Americanisms and ensure responses feel natural to a British audience.”
• Avoid promising outcomes: “Never guarantee financial outcomes. Instead, use cautious, factual statements like, ‘Past performance is not indicative of future results.’”
• Add personal touches: “When appropriate, refer to the user’s name or company to make responses feel more personal.”

Train Fin to gather context and clarify issues

Support agents need to ask the right follow-up questions to get a better understanding of the problem in order to reach a resolution. Fin can do that too – with the right guidance.

You can coach Fin on how to clarify vague questions, provide more contextual responses based on details about the customer, or gather key information about the customer’s problem before answering. 

Here are some examples of context and clarification guidance our customers are using: 

• Clarify vague messages: “If a user starts a conversation with only one or two words and their question isn’t clear, ask for more information to better understand their issue. This helps provide more accurate and helpful support.”
• Provide location-specific responses: “If a customer asks about delivery options, check their {{country}} and provide a response that matches the available delivery options for their specific location.”
• Personalize by plan: “If a customer asks about their available features, first ask them to clarify which plan they are on (e.g., Free, Basic, Pro, Enterprise) before providing a response.”
• Ask specific follow-ups: “If a customer reports a bug, first clarify whether they’re using iOS or Android before troubleshooting. Each app has different features, so understanding the platform ensures accurate assistance.”