RD* studio

Case 02 — Vitta · real estate & financing · at WHF

A paper trail, turned into a conversation

Vitta's employees were buried in manual document collection. Felí guides home buyers through it in a chat — from first interest to subsidy approval.

Role
UX · UI · Service · Research
Scale
The full buying journey — first interest to subsidy approval, on top of Vitta's existing back office
Methods
Design Sprint · Qualitative research · Scrum · LEGO Serious Play
Status
High-fidelity prototype delivered, with a phased implementation roadmap
01 — The situation

A back office that ate the workday

Vitta sells one of the most document-heavy purchases a person ever makes: a home, with financing and subsidy paperwork attached. They already had a back-office system storing client documents and profiles — but it wasn't intuitive. Employees managed collection and organization by hand, chasing files instead of talking to customers.

The mission was to redesign the process end to end: better usability, better efficiency, better customer experience. With one hard constraint — whatever we designed had to integrate with the system they already had. No greenfield.

Time spent chasing paperwork was time not spent on sales and customer interaction.

02 — The turn

Strip the bias first. Then stop designing a form.

Two decisions defined this project — one about how we started, one about what we built.

First: we refused to open with the system. I facilitated a LEGO Serious Play session with Vitta staff from every department and level, so people could think about the problem without their initial biases. Then we framed it formally — true objectives, success and failure criteria, differentiators, risks — and framed the research: what we needed to discover, what hypotheses existed, who could validate them. That framing exposed assumptions that later proved incorrect. Had we designed straight from the brief, we would have built on them.

Second: we stopped treating document collection as a form to be filled and reframed it as a conversation to be had. The concept became a conversational assistant — a traditional messaging UI combined with graphical components for real-estate visualization — designed for modular deployment, so Vitta could roll features out gradually against the legacy back office instead of betting on one big launch.

3customer personas
8 in 8Crazy 8s — UI solutions in eight minutes
2frames before design — challenge + research
4key features in the hi-fi prototype
03 — The work

From workshop bricks to a working assistant

  1. Map who buys, and how the journey actually unfolds

    A second workshop defined three customer personas — demographics, digital behavior, communication preferences — and mapped the customer journey using the "How to Make a Toast" method, extracting from employees how the process really runs from marketing impact to key delivery, not how the org chart says it runs. We then broke the journey into micro-steps and macro-steps, clustered into logical phases that optimized the sequence for clarity and a secure workflow.

    CUSTOMER JOURNEY — MICRO & MACRO STEPS PHASE 01 PHASE 02 PHASE 03 marketing impact key delivery macro-step (milestone / objective) micro-step (small action / task)
    The journey restructured into micro-steps and macro-steps, clustered into logical phases — from marketing impact to key delivery.
  2. Tie the experience to the machine underneath

    Every part of the digital experience was mapped by four elements: content and information; actions (CTAs, external links, interactions); user input fields; and system feedback. In parallel, I worked with Vitta's developers on a technical action map — what data gets collected, how it's stored and retrieved, how system variables behave on the backend — so UX decisions and technical feasibility were never negotiated after the fact.

    EXPERIENCE LAYER — MAPPED BY FOUR ELEMENTS Content information shown Actions CTA · interactions Inputs user data fields Feedback system responses Technical action map — built with Vitta's developers data collected · storage & retrieval · backend variables
    Four-element experience mapping wired to a technical action map — UX and feasibility resolved together, not sequentially.
  3. Sketch fast, validate, then commit to high fidelity

    Ideation ran on hand-drawn low-fidelity prototypes and Crazy 8s — eight UI solutions in eight minutes. The strongest direction, the conversational interface, was refined and validated in low fidelity with internal teams and Vitta's own clients before any pixel-perfect work. Only then did we build the high-fidelity prototype: an assistant that carries a buyer from initial interest in a property all the way to finalized documentation and subsidy approval.

    FELÍ — MESSAGING SPINE + GRAPHICAL MODULES assistant buyer assistant buyer assistant Smart property recommendations real-estate visualization in-thread Document progress tracker submitted vs. pending, at a glance Camera-based submission documents sent from the phone Subsidy validation projected → official amount EACH MODULE = ONE DEPLOYMENT PHASE
    The Felí concept: a messaging UI as the spine, with graphical modules plugged in — each one shippable on its own phase.
    Felí conversation screen from the high-fidelity prototype: the assistant explains the credit analysis can take up to two hours and offers to notify the buyer here or on WhatsApp, with quick-reply buttons. Camera-based document submission screen from the Felí prototype: a CPF card framed for capture with a 'Frente' (front side) overlay and camera controls.
    Straight from the high-fidelity prototype: the conversation doing the paperwork, and the camera flow submitting a document.
04 — Where it landed

Delivered as a system that ships in pieces

Felí was delivered as a high-fidelity prototype with a roadmap for phased implementation: a conversational assistant built to reduce manual work for Vitta's employees, streamline the journey from interest to purchase, and integrate with the existing back office — with each feature deployable gradually rather than in one risky launch. The full trail, from workshop outputs to design system, is documented in Figma.

What this says about how I work

  • I facilitate before I design — LEGO Serious Play and formal framing exposed wrong assumptions before they could shape the product.
  • I treat legacy constraints as design inputs: the technical action map made backend feasibility part of the UX work, not a later negotiation.
  • I design for phased delivery — modular concepts that let a client ship value gradually instead of betting everything on one launch.