Case Study 02

Redesign for Adoption — Transforming the METRO CRM

A CRM used across 17 countries but rarely as intended. The product needed a redesign — but the real challenge was operational: building the research infrastructure, process frameworks, and cross-timezone rituals that would make adoption stick.

Company

METRO.Digital

Role

Experience Designer Consultant — UX Lead, Research Lead, DesignOps

Teams

2 distributed product teams · Germany + India

Timeline

2018 — 2021

5782
SUS Score improved from D (57.3) to B (81.7)
74%
of surveyed users qualified the system as "Easier" to use
17
Countries covered — diverse user profiles and languages
13
Languages supported in the Qualtrics research framework

The challenge

METRO's internal CRM — SAM — was used by sales managers across 17 countries, but adoption was low and satisfaction was rated D (57.3 SUS). The app needed to work across mobile, tablet, and desktop, serve diverse user profiles from Spain to India, and be redesigned while transitioning the entire organization from Angular to React and adopting a new design system used by 25+ products.

The complexity wasn't just the product. It was 20+ stakeholders with misaligned expectations, two product teams owning different business verticals, and a user base that spoke 13 languages and worked in radically different contexts.

"We were designing for a salesperson on the road in Spain and a category manager at a desk in India — same product, completely different needs."

Building the research infrastructure

Before designing anything, I needed to understand what was actually happening. I established a continuous, mixed-method research framework that ran throughout the entire project — not just at the start.

Quantitative — Continuous

Qualtrics In-App Feedback

Implemented Qualtrics across the application, running in all 13 languages. Captured satisfaction scores and recurring pain points at scale — across all countries simultaneously.

Quantitative — Continuous

Web Analytics

Used page view data by device and country to identify where users were spending time, which features were underused, and to validate qualitative findings with behavioral data.

Qualitative — Monthly

User Interviews

In-person and remote interviews conducted every two sprints — understanding goals, pain points, and updating personas as features shipped and the experience evolved.

Qualitative — Targeted

Contextual Interviews

Shadowed users through real tasks — observing unreported pain points that only surface in context. Discovered that users were updating the CRM during appointments, generating duplicate records.

Designing how the team worked

Working with two distributed teams across Germany and India required more than a design process — it required an operational framework. I established the discovery cadence, defined how we shared research, and created the tooling infrastructure that kept both teams aligned.

I created a Design One Pager — a Confluence living document that captured business objectives, user goals, research insights, and technical constraints for each feature before any design began. This became the shared source of truth that prevented misalignment between the two teams.

The design framework was aligned to sprint cadence: visual research and wireframing in week one, PO feedback and client alignment in week two, testing in week three, ready for dev in week four. This gave the client full transparency and managed expectations across a complex stakeholder group.

"Other teams tried to adopt the discovery framework we built — the monthly research cadence, the feedback library, the One Pager template."

Key design decisions

The redesign was grounded in a core insight: SAM users were expert users doing high-frequency tasks over 8-hour days. Every design decision prioritized reduced cognitive load, screen real estate efficiency, and safe experimentation — not visual novelty.

Navigation

Redesigned for frequency

Moved the most-used features — search, customer list, map — to immediately accessible positions. Eliminated multi-tap paths for daily tasks.

Forms

Multi-step → Single flexible form

Collapsed a 2-step activity creation flow into a single flexible form. Reduced friction for the highest-frequency task in the application.

Responsive

Different needs, same product

Mobile prioritized quick, in-context tasks (planning, documenting on the road). Desktop/tablet surfaced planning and analysis features for office work.

Testing

A/B testing with real users

Complex features were tested in A/B format using Figma prototypes. Testing was run in multiple countries — including a multilingual remote setup with interpreters for non-English markets.

B
SUS Score 81.7Up from D (57.3) — a 24-point improvement in system usability
74%
Users said "Easier"of surveyed users qualified the redesign as easier than the previous version
Larger daily engagementRedesign users showed higher daily engagement than users still on the older version

"You don't realize how efficient you're making this team."

— METRO.Digital user, post-launch feedback

Previous

Scaling a Global AI Service

Next

AI Interview Copilot