My approach to design
Jump to section
Overview
Managing color information for digital assets can be time-consuming and unreliable, impacting both administrators and users. Administrators must manually tag color attributes, a process that requires guesswork and limits their ability to focus on other priorities. Meanwhile, users face uncertainty when searching for assets with specific colors, often needing multiple refined searches to find what they need.
What’s the problem?
Role
Product Designer
Tools
Figma, FigJam, Dovetail
Prototype
End-to-end feature
Duration
March 2023 - July 2023
What are we hearing from the people behind the problem?
Our exploration:
How might we automate color tagging to improve efficiency, accuracy, and confidence in retrieving the right assets?
The Solution
To automate color tagging, AI-driven image recognition tools can identify and tag colors in assets, eliminating manual input. This ensures consistent, accurate metadata, enabling users to efficiently search and retrieve assets based on color. This solution enhances workflow efficiency and boosts confidence in asset retrieval.
Improving Asset Search through Automated Color Tagging
Streamlined Workflow: Eliminate manual color tagging by automating the process, saving time and reducing repetitive tasks.
Efficient Asset Discovery: Easily locate images that align with specific brand or project color palettes, enhancing search relevance.
Advanced Search Flexibility: Provide users with the ability to broaden their search with color families or narrow results using precise HEX codes.
Consistent and Accurate Results: Rely on algorithm-driven color detection for precise and consistent color matching, improving overall search accuracy.
The Discovery
I started with empathy mapping using personas, which led to a journey map to identify workflow connections and problem areas. Combined with product requirements, these insights informed task flows. I then conducted a design team workshop with rapid benchmarking, resulting in two concepts for the sprint team. We used a "Like, Wish, Wonder" workshop to refine features for discovery research, focusing on feasibility. After matrix voting, the product manager and I mapped features on the Impact-Effort matrix, refining the design for discovery testing.
Empathy mapping to journey mapping
Must have requirements for the MVP
Key tasks flows for our personas.
Design team workshop, conducting a benchmark on existing color filtering in other platforms.
The initial design concepts before presenting them to the sprint team.
A 'Like, Wish, Wonder' workshop with the sprint team that I conducted.
Insights from the "Like, Wish, Wonder" workshop were synthesized and mapped into features for the next phase of iteration.
Feature voting
Results from the feature voting were mapped into an Impact-Effort matrix to prioritize features by balancing feasibility, desirability, and viability.
Research & Key Findings
Color filter discovery
Approach
Moderated interviews with customers
Recipients, collaborators, contributors
Basic searchers, advanced searchers, and next-level searchers
Objectives and key questions to be answered
Are these preset colors sufficient for users to get the results they need?
Would users need the ability to search with color values other than HEX?
Do users want to filter by more than one color?
Would users benefit from saving their brand colors?
Does this feature help users save time?
Does this feature inspire confidence in users’ searches?
Would users expect to see the color tags in the asset digest?
Color Filter Limitations: Many organizations face challenges with color-based asset search, including the need for HEX code filters or a color palette, and using design tools like Canva for ease.
Search and Metadata Issues: Poor metadata, inconsistent naming, and lack of smart search filters frustrate users, impacting asset discovery.
Feature Improvements: Users suggest incorporating mood-based or vibe-based searches, as well as color prominence sliders and AI-powered features like automatic color detection and contrast-based searching.
User Experience Enhancements: Proposed improvements include custom palettes, better categorization, and smart tag systems.
What we learned
Design Refinement
Introducing an innovative feature that enhances the asset upload process—automated color detection. As images are uploaded, our algorithm identifies the top 8 dominant colors and displays them instantly. These colors are intelligently grouped into families like red, green, and cyan within a 20-degree hue range for more accurate search filtering. For users seeking precision, HEX codes can be input for refined color filtering. This intuitive feature streamlines asset categorization, significantly improving search efficiency and saving valuable time.
Key Learnings & Next Steps
Through thorough planning and research, we successfully crafted an MVP that empowered users to accomplish their core tasks. By being intentional with our discovery process, we were able to validate features initially seen as "delighters" at the MVP stage, which could later be repurposed for upcoming roadmap items, ensuring that we met users' evolving needs over time.
