top of page

WanderBot

A chatbot to help users with trip planning and destination recommendations.

My role

Bot Design, Prototyping, Testing.

Team

Solo project

Timeline

3 months

Focus

Designing a chatbot that understands user requirements.

Tools used

Figma, Voiceflow

Wanderbot hero.png

Project Overview

The WanderBot project emerged from the need to simplify travel planning. Traditional travel planning often involves juggling multiple websites, apps, and resources, creating friction for users who want quick, personalized recommendations.

​

WanderBot provides a streamlined, interactive experience that guides users through the planning process step-by-step. The chatbot asks targeted questions about preferences, budget, and travel style to generate customized itineraries that match individual needs.

Design Process

The design process involved thinking about what kind of a bot WanderBot should be, what should the bot personality be, writing up some sample scripts and gathering training data.​

🤖

Bot description

A chatbot that understands text input and provides a text output.

​

The main use cases - Trip planning and destination exploration.

bot description.png

🤖

Bot personality design

Interaction goals - provide users with personalized and tailored itineraries to plan their travel journeys.

​

Character traits - Focused, Knowledgeable, Responsive, Adaptable, Friendly, Professional.

bot personality.png

Link to the detailed design document - WanderBot Design Process

Testing Protocol

To validate WanderBot's effectiveness, I conducted user testing sessions with potential travelers to evaluate the chatbot's usability, conversation flow, and ability to provide helpful travel recommendations.​

Link to the detailed testing protocol - WanderBot Testing Protocol

Final Design and Conclusion

The final WanderBot design features a conversational interface that guides users through travel planning with natural, context-aware interactions. The bot successfully balances automation with user control, allowing travelers to specify their preferences while receiving personalized recommendations.

​

Key design elements include clear conversation flows, helpful prompts for user input, and transparent communication about the bot's capabilities.

Link to the WanderBot prototype - WanderBot Prototype

Key Learnings

  • I had initially designed a bot that had no turn taking sequence. If the bot is not a LLM it is very important to implement turn taking. It makes it easy for the users to understand when the bot is done answering. Also, it makes it easy for the users to just answer the question the bot posts.

  • Designing bot personalities takes research and effort. How the bot responds heavily depends on its personality and it makes or breaks the conversation. Also, the bot personality needs to be aligned with the brand identity.

Next Steps

  • Implement button-based interaction options alongside conversational interface to provide users with clear action choices and reduce input friction.

  • Expand utterance coverage across all intents to capture diverse user expressions and improve natural language understanding accuracy.

  • Add comprehensive safety information features including destination safety ratings, health advisories, and emergency contact resources.

  • Integrate connectivity guidance covering SIM card options, data plans, WiFi availability, and communication solutions for different countries.

  • Incorporate rich media responses including destination images, interactive 3D map views, and curated links for deeper exploration of recommended locations.

bottom of page