Banco Falabella — Lía Chatbot

Context

Banco Falabella needed a conversational channel for four financial businesses: Banco, CMR, Viajes, and Seguros. Each line had different services and expectations, but customers needed one simple, friendly assistant.

The hard part wasn't writing good responses; it was building a structure that guided users and helped the assistant answer accurately across very different kinds of requests.

My role

I worked as UX Specialist with a Conversational UX focus: I structured the assistant's logic, defined guided flows, trained intents, and helped create a methodology for improving the chatbot over time.

• Designed conversational flows for banking, CMR, travel, and insurance use cases.
• Structured intents, templates, and guided interaction patterns.
• Helped train Lía by identifying what information the chatbot needed and what could create confusion.
• Worked with business stakeholders to understand services and expected requests.
• Designed visual assets and interface components for the chatbot's UI.

The challenge

The main challenge was designing a chatbot that could support multiple financial services without overwhelming users or breaking the assistant's response quality.

• User challenge: users needed quick answers, but open-ended questions led to confusion and dead ends.
• Business challenge: one assistant had to serve four business lines with a consistent experience.
• Design challenge: overtraining created poor responses, so the experience needed clear structure and well-organized intents.

Goals & success criteria

Process

01 — Discover

Understanding how users ask for help in each financial service

I first understood the business lines and the kind of support users needed from each one: where conversations became ambiguous and which topics needed controlled flows instead of open-ended questions.

• Identified recurring requests, inquiries, and transactional needs.
• Mapped where open-ended conversations could create misunderstanding.
• Analyzed how information should be grouped to make the chatbot easier to train and easier to use.

Early discovery structure used to understand how different financial services could live inside one conversational assistant.

02 — Define

Turning scattered services into a conversational architecture

Then I defined the structure behind the assistant: organizing intents, deciding when to use templates, and shaping a model where users choose options instead of describing everything from scratch.

• Grouped intents by business line, user need, and expected outcome.
• Defined guided paths for frequent inquiries and transactions.
• Helped define how Lía should behave when she did not understand a request.

Intent architecture used to organize multiple financial services into a single conversational model.

03 — Develop

Designing guided flows instead of fully open conversations

The most important decision was not to treat the chatbot as a fully open question-and-answer system: too much freedom made the experience less reliable. Guided flows and templates reduced the risk of wrong answers.

• Designed conversational flows for key use cases.
• Created guided templates to help users choose the right path.
• Adjusted flows when training results showed confusion or wrong responses.

Guided flow example showing how structured options reduced ambiguity compared with open-ended questions.

04 — Deliver

Launching a friendlier, more scalable assistant

The result was an assistant that could serve several financial businesses with a friendly, structured experience, designed to improve over time through more intentional training.

• Delivered conversational flows, templates, and intent structures.
• Improved the quality of interactions by reducing overtraining and ambiguous inputs.
• Built a reusable Conversational UX foundation for future improvements.

Final conversational experience simplified for portfolio purposes to show structure without exposing sensitive business logic.

Before / after

Before-and-after comparison of the conversational model, showing how structured paths replaced the open-ended input that produced failed responses.

Final solution

The proposal focused on three improvements:

Summary of the final solution, where the three pillars — guided flow, intent architecture, and a friendlier tone — operate within a single conversation.

Impact & results

Obstacles & trade-offs

The balance was between user freedom, business coverage, and chatbot reliability, in a regulated environment that required coordinating every flow with legal and compliance. The strong solution was structuring the experience better, not adding more content.

• Avoiding overtraining while still covering multiple business needs.
• Balancing guided templates with the user's expectation of natural conversation.
• Maintaining a friendly tone while handling financial information.

What I learned

This project shaped the way I think about UX, well beyond chatbots: structure is what allows a product to scale.

• Information architecture is critical in conversational experiences.
• Overtraining a chatbot can make the experience worse, not better.
• Guided flows can create more confidence than fully open input.