Data Tag Management
I designed a new tagging system to help enterprise users classify and take action on millions of files. The project focused on solving problems of scale, security, and role-based access. It turned a messy, unstructured system into one that is organized, governed, and easy to use.
Type
Web App
Enterprise UX
Interaction Design
Role
UX Designer
Team
1 Product Manager
3 Software Developers
Duration
4 months
PROBLEM
From chaos to clarity: a movie studio’s challenge
Imagine a movie production studio managing terabytes of files. These include raw footage, voiceovers, wireframes, concept art, and marketing assets, etc. All of them are scattered across different file types and storage systems.
The studio wanted to organize these assets for faster search and smarter actions.
They had requests like:
“Find all French-dubbed Shrek audio files.”
“Move VHS footage of Frozen to cold storage.”
“Tag all storyboards for Toy Story with one click.”
That’s when we realized that our platform needed a powerful, scalable, role-based tagging system.
It wasn’t just about classifying files. It had to support fast actions, smart search, and access control.
EXPLORATION
Organizing massive data safely and efficiently
Tagging wasn’t just about labeling files. It was a step toward making unstructured data more manageable, secure, and actionable.
To explore how this could work within Komprise, I broke down:
Who would manage or use tags (IT admins vs regular users)
What actions those roles needed to take
Where those actions would occur across the product
These insights helped form the foundation for a role-based tagging system optimized for scale, security, and operational clarity.
INFORMATION ARCHITECTURE
Mapping the tagging ecosystem
Tagging at Komprise wasn’t just a feature. It needed a clear, scalable architecture to support real-world workflows.
To support tagging across the platform, I defined four core UI components, each with a distinct role:
Tag Library
A centralized location for users to view, create, edit, and delete tags.Tagging Jobs
Tagging operations can take time. This history panel tracks tag-related jobs such as applying or untagging, allowing users to monitor ongoing activity.Apply Tag
A small contextual popup for applying tags to specific data. Users can select from existing tags in the library and apply them to filtered results.Un-tag
A counterpart to Apply Tag. It allows users to remove existing tags from selected items. Tags are removed from the data, but remain in the Tag Library.
Where these features live matters.
Tag Library and Tagging Jobs are global tools for managing and tracking tags, so they are located in Global Settings.
Apply Tag and Un-tag appear in the context of specific user actions, embedded directly in task-based views.
Currently, they are integrated into *Deep Analytics and *Data Stores.
This role-based architecture and intentional placement ensured tagging was both accessible and operationally intuitive regardless of dataset scale or user expertise.
*Deep Analytics: Where users can filter data into query. *Data stores: Where users can view data structure and details.
FINAL DESIGN 01
Tag library: the central hub
I designed the Tag Library to serve as a familiar yet powerful control center. Instead of reinventing layouts, I reused common components from across the platform, minimizing cognitive load. Most users create a few keys, but many values — so I prioritized keys as anchors, using accordion rows to show relevant values. Finally, role-based visibility ensured that only users with proper permissions could view or modify tag content.
FINAL DESIGN 02
Create tags: scalable input without errors
Creating multiple tags needed to be fast — but not error-prone. I designed the input form to accept comma-separated values and immediately show them as pills, previewing what would be created. This gave users a visual checkpoint to confirm entries before finalizing. I also added protections like character limits, duplicate prevention, and validation feedback to ensure input accuracy.
FINAL DESIGN 03
Apply tags: seamless and contextual
I reused the visual design of the “Create Tag” popup — so users wouldn’t need to relearn anything. Instead of text input, I used dropdowns that double as searchable inputs, reducing friction for tag application. This lightweight interaction supported both scanning and speed — users could visually pick or type to filter. The result: a consistent and efficient tagging experience.
FINAL DESIGN 04
Un-tag: quick undo without confusion
Un-tagging needed to be fast and low-risk. I surfaced a tag list popup directly from the data context (file or query), so users could instantly see what was applied. Checkboxes allowed bulk selection and action, with clear previews and minimal clicks. I made sure this didn’t remove the tag from the library — just disassociated it from the selected data.
