HICX · Supplier management
Data quality.
A clear path to resolution.
Designing a central workspace for HICX that connects supplier data health with a clear path to correction.
HICX · UX / PRODUCT DESIGNTL;DR
- Role
- UX / Product Designer · interaction & workflow design
- Team
- HICX · Supplier management
- Timeline
- During my HICX tenure · 2025–2026
- Result
- A unified workspace for data health, investigation and correction.
Bad data was visible in spreadsheets.
Its health was invisible in the platform.
Supplier records contained many data quality issues. Users had no live view of data health within the platform to tell them what they could trust.
Maintaining healthy data is an ongoing challenge in supplier management. At HICX, the problem extended beyond individual incorrect values: users needed to understand the quality of each supplier’s data and know which issues required attention.
What supplier interviews revealed
Across numerous supplier interviews and usability testing sessions, participants explained that resolving bad data took considerable time. Existing processes involved large Excel files with many rows, making the work of reviewing and correcting supplier information cumbersome.
Users were also unsure whether the data was reliable. There was no live indicator within the platform showing how healthy supplier data was, so they lacked a clear view of its current quality.
The problem to solve
Bring data health and issue resolution into one workspace: make problems visible, help users prioritise what needs attention and provide a clear route to investigate and correct each issue. The aim was to reduce the effort of managing corrections through large spreadsheets and give users greater confidence in the data.
Turn a spreadsheet task
into a resolution workflow.
The research pointed to two connected needs: understand the state of supplier data and act on the issues behind it. The design brings both into the platform through four steps.
Separate the overview
from the work queue.
The dashboard answers “How healthy is our data?” through issue volume, severity and resolution activity. The issues table answers “What should I work on next?” by bringing individual issues, ownership and status into a focused queue. Separating these views connects overall data health with the work needed to improve it.
Make the issue the unit of work.
A supplier can have several unrelated data problems. Making each issue a row gives every correction its own severity, owner and status. Users can focus on a specific problem without treating the entire supplier record as one unresolved task.
Investigate without
losing the queue.
Large spreadsheets force users to find the relevant information among many rows. The detail panel brings the selected issue’s supplier, affected field, current value, ownership and audit history into one place while keeping the queue in view. Users can investigate a problem without losing their position in the wider workload.
Make the consequence of saving explicit.
The correction dialog compares the current and new values and explains what happens after saving: the rule is evaluated, and a passing result closes the issue and records the change. This makes the resolution sequence explicit, so users can distinguish submitting a correction from confirming that it satisfies the rule.
Suggest a correction.
Keep the decision understandable.
The assisted-resolution concept addresses the effort of researching a correction. It pairs a suggested fix with supporting evidence and a preview of the change. Where several values are plausible, users can compare alternatives or choose a manual correction. The design keeps the recommendation reviewable and the decision with the user.
| Design challenge | Approach illustrated |
|---|---|
| Understand the recommendation | Show the source, explanation and field-level difference. |
| Handle ambiguity | Offer alternatives and a manual correction path rather than implying a single certain answer. |
| Keep actions accountable | Validate the correction, show the committed value and retain an audit history. |
From evidence to confirmation.
Explore the interactive screen walkthrough: review a tax-format issue, inspect the suggested correction and see the confirmation state.
Compare the rejected value with the expected format, then inspect the proposed correction before applying it.
Make data health actionable.
Knowing that data is wrong is only useful when people can understand the issue and act on it.
The hub connects a data-health overview with an issue queue, contextual investigation and correction. Suggestions bring evidence into the same decision.
The central design principle is to make every step understandable: what needs attention, why a change is suggested and what happens when it is applied.