For effective bots, follow these best practices for conversational design:

  • Develop a voice: Create a personality that identifies with your brand and that consumers can connect with.
  • Identify as a bot: Be transparent by using terms like "virtual assistant" or "automated assistant."
  • Be transparent about Generative AI: If you're using Generative AI, inform the end user via a bot message that info provided via the bot is generated using cutting-edge Generative AI technologies and, as such, might be prone to inaccuracy from time to time. This type of notice is an industry standard and a best practice based on the compliance requirements in some areas (e.g., in some U.S. states).
  • Explain capabilities: Detail up front what the bot is able to help with to set consumer expectations.
  • Use contractions: "I'm" sounds more natural than "I am" and reflects how humans actually speak.
  • Avoid ambiguity: When a bot doesn't recognize an intent, it's recommended to have a disambiguation response.
  • Short and simple: Use short dialogue to keep consumers engaged and simple words everyone can understand.
  • Visualization: Line breaks and separate messages can break up long text; emojis and rich content can help to liven up the conversation.
  • Consider context: Returning consumers can be greeted differently than when they entered the experience for the first time.
  • Tango: Allow consumers to be connected with an agent at their request, and when the bot doesn't understand after 2 attempts.
  • Validate intent: To avoid consumer frustration, the bot should confirm the recognized intent before moving down a flow.