All cases
Case Study 03 Hackasong MX 2025

4Health Habit Tracker

Client
Accenture Song
My role
Product Design Lead
Timeline
Oct 2025 · 48 hrs
Phase
Hackasong MX
The challenge

Build a health app concept in 48 hours — with AI at every step

48
Hours from blank canvas to award-winning prototype
3
Designers on the team across all FORM phases
4
AI tools
Used across research, synthesis, PRD and prototype
1
Hackasong MX award won at Accenture Song's internal hackathon
My contribution
Product Design Lead
  • Led the team across all three FORM phases during the 48-hour sprint
  • Directed AI-assisted research using Microsoft Copilot for desk research and competitive analysis
  • Generated synthetic user personas with an internal Accenture generative AI tool
  • Authored the PRD and refined prompts with Copilot to accelerate definition
  • Built the high-fidelity interactive prototype entirely in Figma Make (AI-generated)
  • Presented the final concept to Accenture Song judges and won the Hackasong MX award
Context

Hackasong MX is Accenture Song Mexico's internal design hackathon — a 48-hour competitive sprint where teams race to design, prototype and pitch an app concept from zero. The 2025 edition challenged teams to combine human-centered design with AI tooling.

4Health tackles a real behavioral challenge: people struggle to build and maintain health habits consistently. The app addresses this through habit creation, streak tracking, daily check-ins and progress visualization — all wrapped in a low-friction mobile experience.

What made this sprint different was the deliberate application of AI tools across every phase of the FORM methodology — compressing what typically takes weeks of discovery into hours, without compromising rigor. The approach earned the team the Hackasong MX award.

Phase 01
Discover & Describe
AI-Accelerated Research
Phase 02
Design & Implement
Rapid Design with AI
Phase 03
Validate & Ship
Hackathon Presentation
01
Discover & Describe
AI-Accelerated Research & Problem Definition
Hours 0 – 16 · Oct 2025

With only 48 hours total, traditional discovery was not viable. Instead, we ran an AI-augmented research sprint: Copilot for desk research and competitive benchmarking, an internal Accenture generative AI tool for synthetic personas, and Copilot again to draft and refine the PRD — giving us a solid strategic foundation in under half a day.

AI tool stack — Phase 01
01
Desk Research
Microsoft Copilot
Competitive analysis of habit tracking apps: behavioral patterns, key friction points, feature benchmarking.
02
Synthetic Personas
Accenture Internal AI
Generated representative user personas from behavioral data without primary research recruitment.
03
PRD Definition
Microsoft Copilot
Drafted product requirements document and refined prompts to align team around scope and success criteria.
The problem
Habit Formation is Hard

Research confirmed the core challenge: most people abandon health habits within the first two weeks. The root causes are lack of immediate feedback, unclear progress indicators, and low commitment mechanisms. 4Health was designed to address each of these directly through streaks, daily check-ins and visual progress.

Target users (synthetic personas)
Two Core Profiles

The Restarter — 28–35 years old. Motivated to build healthy habits but has failed before. Needs clear wins early on and gentle accountability.

The Builder — 22–30 years old. Consistent exerciser wanting to add non-physical habits (sleep, hydration, mindfulness). Needs a unified tracker with minimal friction.

Phase 01 — Key deliverables (click to preview)
Research Brief — Desk Research & Competitive Analysis
Research · Copilot
Synthetic User Personas — The Restarter & The Builder
Personas · Accenture AI
Product Requirements Document (PRD)
PRD · Copilot
02
Design & Implement
Rapid Design with Figma Make
Hours 16 – 40 · Oct 2025

Armed with the PRD and personas, we moved directly into high-fidelity design using Figma Make — Figma's AI-powered prototyping tool. Rather than building screen-by-screen, Figma Make generated interactive flows from our prompts, which we then refined and iterated. The result: a fully functional, high-fidelity prototype in a fraction of traditional design time.

Core feature set
Habit Creation
Guided flow to define new habits — frequency, category, reminder time and goal target. Minimal fields, maximum clarity.
Streak Tracking
Visual streak counters that reinforce consistency. Missing a day breaks the streak — creating a powerful behavioral commitment mechanism.
Daily Check-ins
One-tap daily habit completion from the home screen. Friction-free logging to maximize the chance of the behavior becoming automatic.
Progress Visualization
Weekly and monthly progress views. Heatmaps and trend lines that make growth visible and provide motivational context.
The tool
Figma Make

Figma Make is an AI-powered feature within Figma that generates interactive prototypes from natural language prompts. We used it to go from wireframe sketches to a navigable, high-fidelity prototype — with real interactions, transitions and component consistency — without writing a single line of traditional design spec.

Why it worked
Prompt as Design Brief

The PRD built in Phase 01 became the direct input for Figma Make prompts. This tight loop between research output and design generation eliminated the traditional handoff delay and kept the team focused on refining and validating rather than building from scratch.

Phase 02 — Key deliverables (click to preview)
Figma Make Prototype — Full Interaction Flows
High-fidelity · Figma Make
03
Validate & Ship
Hackathon Presentation & Award
Hours 40 – 48 · Oct 2025

The final phase of the sprint was dedicated to packaging, storytelling and presenting to the Accenture Song judging panel. We structured the pitch around the FORM methodology — showing how AI tools amplified each phase — and demonstrated the interactive prototype live. The result was the Hackasong MX 2025 award.

Award won
Hackasong MX 2025
Awarded by Accenture Song Mexico for innovation in AI-augmented design methodology and quality of the high-fidelity prototype delivered within the 48-hour sprint. Judged by senior leadership from Accenture Song Mexico.
Key learnings from the sprint
01
AI Compresses Time
AI tools collapsed the discovery-to-prototype timeline from weeks to hours — without sacrificing the rigor that judges care about.
02
PRD as Prompt
The tightest loop was between PRD and Figma Make. A well-structured brief became a design generator — a workflow worth formalizing.
03
FORM Still Leads
AI did not replace methodology — it accelerated it. FORM provided the structure; AI provided the speed. That combination is the real differentiator.
Outcomes & impact

48 hours.
Award-winning.

1st
Place at Hackasong MX 2025 — Accenture Song Mexico's internal design hackathon
48h
Research-to-prototype in under 48 hours — from blank canvas to award-winning interactive demo
4
AI tools integrated across every phase: Copilot, Accenture internal AI, and Figma Make
3
Designers on the team — demonstrating that a small AI-augmented team can outperform larger traditional squads
AI augments, not replaces, methodology
The hackathon proved that FORM is stronger with AI, not redundant. The methodology provided the strategic skeleton; AI tools filled each phase with speed and precision. Designers who master this combination become a category of one.
Great briefs make great prototypes
The quality of Figma Make output was directly proportional to the quality of the PRD. Investing the first 16 hours in rigorous (AI-assisted) definition meant the design phase was generation and refinement, not exploration from zero.
Synthetic research is real research
AI-generated personas are not a shortcut — they are a different research modality. When grounded in desk research and behavioral principles, synthetic personas provide directionally valid insights at a fraction of the time cost of primary recruitment.
Back to all cases
Next case
Santander Digital Payroll