Go Back
Order Management System
Order Management System
Order Management System
The Order Management System (OMS) was a mission-critical platform, driving over 50% of company revenue. Agents used it to create orders live on customer calls, where speed and accuracy were essential. The legacy system was slow, error-prone, and couldn’t support multiple products in a single order, forcing agents to repeat flows and extend calls.
This initiative was part of a broader digital transformation to redesign a legacy CRM used by 2,000+ agents across 16+ countries. The goal was to streamline workflows, improve usability, and speed up customer order processing for a multilingual workforce.
During the company’s digital transformation, I stepped in as the sole designer, owning the redesign under tight timelines and limited developer capacity. To capture real pain points, I conducted 10+ agent interviews structured around live workflows, ensuring the new design matched the pace and pressure of on-call order creation.
The image below shows the legacy OMS interface before redesign — a cluttered, scroll-heavy layout that slowed agents down and made order creation error-prone
The Order Management System (OMS) was a mission-critical platform, driving over 50% of company revenue. Agents used it to create orders live on customer calls, where speed and accuracy were essential. The legacy system was slow, error-prone, and couldn’t support multiple products in a single order, forcing agents to repeat flows and extend calls.
This initiative was part of a broader digital transformation to redesign a legacy CRM used by 2,000+ agents across 16+ countries. The goal was to streamline workflows, improve usability, and speed up customer order processing for a multilingual workforce.
During the company’s digital transformation, I stepped in as the sole designer, owning the redesign under tight timelines and limited developer capacity. To capture real pain points, I conducted 10+ agent interviews structured around live workflows, ensuring the new design matched the pace and pressure of on-call order creation.
The image below shows the legacy OMS interface before redesign — a cluttered, scroll-heavy layout that slowed agents down and made order creation error-prone
The Order Management System (OMS) was a mission-critical platform, driving over 50% of company revenue. Agents used it to create orders live on customer calls, where speed and accuracy were essential. The legacy system was slow, error-prone, and couldn’t support multiple products in a single order, forcing agents to repeat flows and extend calls.
This initiative was part of a broader digital transformation to redesign a legacy CRM used by 2,000+ agents across 16+ countries. The goal was to streamline workflows, improve usability, and speed up customer order processing for a multilingual workforce.
During the company’s digital transformation, I stepped in as the sole designer, owning the redesign under tight timelines and limited developer capacity. To capture real pain points, I conducted 10+ agent interviews structured around live workflows, ensuring the new design matched the pace and pressure of on-call order creation.
The image below shows the legacy OMS interface before redesign — a cluttered, scroll-heavy layout that slowed agents down and made order creation error-prone
The Order Management System (OMS) was a mission-critical platform, driving over 50% of company revenue. Agents used it to create orders live on customer calls, where speed and accuracy were essential. The legacy system was slow, error-prone, and couldn’t support multiple products in a single order, forcing agents to repeat flows and extend calls.
This initiative was part of a broader digital transformation to redesign a legacy CRM used by 2,000+ agents across 16+ countries. The goal was to streamline workflows, improve usability, and speed up customer order processing for a multilingual workforce.
During the company’s digital transformation, I stepped in as the sole designer, owning the redesign under tight timelines and limited developer capacity. To capture real pain points, I conducted 10+ agent interviews structured around live workflows, ensuring the new design matched the pace and pressure of on-call order creation.
The image below shows the legacy OMS interface before redesign — a cluttered, scroll-heavy layout that slowed agents down and made order creation error-prone
Employer:
Employer:
Cimpress
Role:
UX Designer
Year:
2025
2023
Research & Discovery
Interviewed 10 agents across different geographies, roles, ages, and tenures to capture a wide perspective.
Partnered with PMs and senior stakeholders to map workflow breakdowns and align design goals.
Quantity selection varied across product types, increasing cognitive load and slowing agents down.
Splitting orders meant repeating the full Add to Cart process multiple times, wasting time and creating opportunities for error.
Linked pain points to agent KPIs such as average call duration, conversion rates, and orders processed per day, showing the direct impact of UX inefficiencies on performance.
We also applied Service Blueprinting to map the customer–agent–system relationship end to end.
This revealed:
Disconnects between what the system showed vs. what agents needed
Manual workarounds and fallback behaviours
Low trust in data, leading to repeated confirmations
Time lost in search and verification, scaling into hidden operational costs
Research & Discovery
Interviewed 10 agents across different geographies, roles, ages, and tenures to capture a wide perspective.
Partnered with PMs and senior stakeholders to map workflow breakdowns and align design goals.
Quantity selection varied across product types, increasing cognitive load and slowing agents down.
Splitting orders meant repeating the full Add to Cart process multiple times, wasting time and creating opportunities for error.
Linked pain points to agent KPIs such as average call duration, conversion rates, and orders processed per day, showing the direct impact of UX inefficiencies on performance.
We also applied Service Blueprinting to map the customer–agent–system relationship end to end.
This revealed:
Disconnects between what the system showed vs. what agents needed
Manual workarounds and fallback behaviours
Low trust in data, leading to repeated confirmations
Time lost in search and verification, scaling into hidden operational costs
Research & Discovery
Interviewed 10 agents across different geographies, roles, ages, and tenures to capture a wide perspective.
Partnered with PMs and senior stakeholders to map workflow breakdowns and align design goals.
Quantity selection varied across product types, increasing cognitive load and slowing agents down.
Splitting orders meant repeating the full Add to Cart process multiple times, wasting time and creating opportunities for error.
Linked pain points to agent KPIs such as average call duration, conversion rates, and orders processed per day, showing the direct impact of UX inefficiencies on performance.
We also applied Service Blueprinting to map the customer–agent–system relationship end to end.
This revealed:
Disconnects between what the system showed vs. what agents needed
Manual workarounds and fallback behaviours
Low trust in data, leading to repeated confirmations
Time lost in search and verification, scaling into hidden operational costs
Research & Discovery
Interviewed 10 agents across different geographies, roles, ages, and tenures to capture a wide perspective.
Partnered with PMs and senior stakeholders to map workflow breakdowns and align design goals.
Quantity selection varied across product types, increasing cognitive load and slowing agents down.
Splitting orders meant repeating the full Add to Cart process multiple times, wasting time and creating opportunities for error.
Linked pain points to agent KPIs such as average call duration, conversion rates, and orders processed per day, showing the direct impact of UX inefficiencies on performance.
We also applied Service Blueprinting to map the customer–agent–system relationship end to end.
This revealed:
Disconnects between what the system showed vs. what agents needed
Manual workarounds and fallback behaviours
Low trust in data, leading to repeated confirmations
Time lost in search and verification, scaling into hidden operational costs
Design Approach
Consistency in Quantity Selection: Standardized flows across products so agents always entered quantity upfront, reducing mental switching.
Bulk Uploads: For large, multi-variant orders, introduced an Excel upload flow, letting agents handle variations faster and with fewer errors.
Layout Exploration: Compared single-page vs. multi-step flows:
Prototyping & Testing: Built rapid prototypes to test with agents under real workflows. Quick validation cycles ensured designs aligned with their mental models while staying feasible for developers.
Design Approach
Consistency in Quantity Selection: Standardized flows across products so agents always entered quantity upfront, reducing mental switching.
Bulk Uploads: For large, multi-variant orders, introduced an Excel upload flow, letting agents handle variations faster and with fewer errors.
Layout Exploration: Compared single-page vs. multi-step flows:
Prototyping & Testing: Built rapid prototypes to test with agents under real workflows. Quick validation cycles ensured designs aligned with their mental models while staying feasible for developers.
Design Approach
Consistency in Quantity Selection: Standardized flows across products so agents always entered quantity upfront, reducing mental switching.
Bulk Uploads: For large, multi-variant orders, introduced an Excel upload flow, letting agents handle variations faster and with fewer errors.
Layout Exploration: Compared single-page vs. multi-step flows:
Prototyping & Testing: Built rapid prototypes to test with agents under real workflows. Quick validation cycles ensured designs aligned with their mental models while staying feasible for developers.
Design Approach
Consistency in Quantity Selection: Standardized flows across products so agents always entered quantity upfront, reducing mental switching.
Bulk Uploads: For large, multi-variant orders, introduced an Excel upload flow, letting agents handle variations faster and with fewer errors.
Layout Exploration: Compared single-page vs. multi-step flows:
Prototyping & Testing: Built rapid prototypes to test with agents under real workflows. Quick validation cycles ensured designs aligned with their mental models while staying feasible for developers.
Impact
Reduced average handling time per order by eliminating redundant steps.
Enabled agents to process more orders per day, improving throughput.
Lowered error rates by enforcing structured, guided workflows.
Supported the company’s digital transformation, aligning the design with React migration and scalability needs
Impact
Reduced average handling time per order by eliminating redundant steps.
Enabled agents to process more orders per day, improving throughput.
Lowered error rates by enforcing structured, guided workflows.
Supported the company’s digital transformation, aligning the design with React migration and scalability needs
