Retail Operations · Product Design · 2024 · Shipped
Product Designer
Grupo Boticário.
store experience.
VMEX — the platform that replaced Grupo Boticário's manual visual-merchandising journey, cutting franchise support by 45% (≈ R$140k) and bringing precision to how store pieces are briefed, produced and distributed each cycle.
¶ Snapshot
Client
Grupo Boticário
Year
2024
Role
Product Designer
Duration
Oct 2022 – Dec 2024 (~2 years)
Team
Cross-functional: Product, Engineering & Business
Platform
Web · in-store platform
Status
Shipped
Scope
UX Research · Product Design · Usability Testing
My contribution
Owned the end-to-end Product Design: Discovery and UX Research, interviews and customer-journey mapping, Information Architecture, user flows, wireframes, high-fidelity prototypes in Figma, and usability testing to validate solutions before implementation. Worked in an Agile, cross-functional setup with Product, Engineering and Business — helping prioritise opportunities and aligning with stakeholders through a data-driven approach.
¶ Problem
What was broken
Context
Grupo Boticário's store-experience journey — briefing, producing and distributing visual-merchandising (VM) pieces to franchises every cycle — ran on a manual spreadsheet. Distribution errors, mid-cycle briefing changes and manual picking made the routine fragile and generated heavy franchise support.
Baseline
Manual VM spreadsheet · distribution & quantity errors · budget rework turning 2 steps into 6 · high franchise contact rate
If we did nothing
Every manual step rippled into wrong piece counts, late production and boxes left open waiting on delayed pieces — and into franchises calling for support at every cycle turn.
¶ Constraints
- 01A multi-store operation: the solution had to standardise processes without breaking what local teams already did well.
- 02Replacing entrenched manual routines that store teams used every day — adoption had to feel like relief, not extra work.
- 03Delivered iteratively in an Agile setup, in lockstep with Product, Engineering and Business.
¶ Process
How we worked
Research
A qualitative, attitudinal study: 21 in-depth individual interviews across internal teams, suppliers and franchisees — 100% of them people who work day-to-day in the VM spreadsheet. The journey was mapped across three pillars (Architecture, VM and Digital Exposure) to locate the real pains and set the priority for an MVP. From there I defined the IA and user flows, designed wireframes and hi-fi prototypes in Figma, and ran usability testing before build.
Hypotheses
- H1A single, user-centered platform will remove the manual workarounds store teams rely on.
- H2Standardising operational flows will cut support contacts more than training alone.
- H3Grounding prioritisation in usage data and feedback will keep the roadmap on the pains that actually move efficiency.
¶ Decisions
Each choice listed beside the alternative we deliberately rejected — and why.
D01
We chose
Replace manual, ad-hoc steps with one standardised digital flow.
Over
Digitise each store's existing process as-is.
Because
Paving the cow paths would have locked in the inconsistency that drove support contacts. Standardisation was the whole point of the initiative.
D02
We chose
Validate with usability testing before implementation.
Over
Ship first and learn in production.
Because
Store teams couldn't absorb churn in their daily routine; testing de-risked rollout and protected adoption.
D03
We chose
Prioritise with a data-driven approach — feedback plus metrics.
Over
Prioritise purely by stakeholder request volume.
Because
Metrics kept the roadmap honest, focused on the pains that actually moved contact rate and operational efficiency.
¶ Craft
¶ By the numbers
What the research drew on, and what the platform moved.
21
in-depth interviews — internal teams, suppliers & franchisees
100%
of them work day-to-day in the VM spreadsheet
−45%
franchise contact rate (the OKR target was −30%)
R$140k
value reflected by the contact-rate drop
¶ The design process
End to end, from discovery to a solution validated with real users before it ever shipped.
Discovery & UX Research
Interviews & customer journey
Information architecture
User flows & wireframes
Hi-fi prototypes in Figma
Usability testing
¶ Key discoveries
From the interviews, the operation's friction concentrated in three places.
Manual piece distribution generates errors in how many pieces get produced — distribution is also handled by the Creative team for cycle-specific pieces.
Briefing, quantity and description changes land during budget phases 1 & 2 — turning what should be 2 steps into 6 (first round R$200k, second R$2M).
Production time shrinks every cycle: graphic suppliers deliver in phases, and Antilhas' manual picking generates piece errors — boxes are bip-scanned, not each piece.
¶ Attributes for the value proposition
Research framed what VMEX had to deliver — functionally and emotionally.
Functional
- Stop being a manual process
- Strong usability
- Precise information
- Agile piece distribution
- Unify the journey into one system
Emotional
- Relationship & proximity across journey areas
- Tranquility for internal teams to work precisely
- Clear benefits & differentials
- Minimize impact for the franchisee in VM
¶ Customer journey
The end-to-end VM journey mapped in Miro — the shared artefact used to align Product, Engineering and Business around the real operational pains.
¶ The prototype — navigable
The VMEX monthly-cycle flow running live inside this page — click through it just as you would in the product.
¶ Where it's headed — OKRs 2024
- MVP launched Dec 2023 — first CSAT round completed Jan 2024
- Automate manual steps to cut distribution errors across Boti cycles
- Expand VMEX to other group brands — QDB, Eudora and more
- Pilot barcode-tagged pieces (codes already with Antilhas)
- Tactical OKR: −30% contact rate at cycle turn — achieved −45%
¶ Outcome
−45%
Contact rate
≈ R$140k reflected
VMEX platform
VM process
Dec 2023
MVP shipped
first CSAT round Jan 2024
Qualitative signal
Replacing the manual VM spreadsheet with VMEX gave internal teams, suppliers and franchisees one precise source of truth — less rework in briefing and distribution, fewer wrong piece counts, and a calmer cycle for everyone with an interface in the journey.
Business impact
Cut franchise contact rate by 45% — a ≈ R$140k impact — and set VMEX up to expand to other Grupo Boticário brands (QDB, Eudora) and to a barcode-tagged piece pilot.
¶ Reflection
What I'd do differently
Across two years, the hardest part wasn't the interface — it was standardising a journey shared by internal teams, suppliers and franchises without erasing what each already did well. Interviewing the people who live in the spreadsheet, then validating before rollout, is what made VMEX actually stick.
Next case · 03 / 04
Open ↗Mercado Livre
meli play.
