Strategic UX Redesign
SEO AI
SEO AI helps grow traffic, boost ROI, increase LLM visibility, and cut SEO costs
Year :
2025-2026
Industry :
Digital Marketing SaaS / SEO Technology
Client :
Search Atlas
Project Duration :
2 weeks
Tools:
Figma, Claude.Ai, Figma Make, Nielsen Heuristics
Search Atlas provides enterprise-grade SEO AI OTTO for digital marketers. Their Keyword Rank Tracker is a critical feature used daily by SEO professionals to monitor hundreds of keywords across multiple search engines. However, user feedback indicated frustration with the interface's complexity and difficulty completing routine tasks efficiently.
As a Senior UI/UX designer....
How might I redesign the Keyword Rank Tracker to reduce cognitive load and improve task completion speed for SEO professionals managing large keyword portfolios, while maintaining access to comprehensive data?
Problem
The Challenge
The current dashboard presents 11 simultaneous columns with 15 elements per row, creating cognitive paralysis that overwhelms users instead of guiding them toward informed decisions.
Core Finding
The inter face prioritizes data completeness over usability, requiring a strategic shift toward progressive disclosure
• 10 Critical Issues Identified
• 8 Heuristics Violated • 4 High Priority Fixes
Process
Strategic Approach
I employed a systematic heuristic evaluation framework combined with AI-assisted analysis to maximize efficiency and thoroughness:
2. AI-Assisted Analysis with Claude.ai
This approach demonstrates strategic tool selection — using AI not as a replacement for design expertise, but as a productivity multiplier that allowed me to focus cognitive energy on strategic decisions rather than tactical execution.
Leveraging Claude.ai as a design thinking partner accelerated my analysis process
• Rapid iteration on solution concepts
• Generation of multiple design alternatives
• Documentation and presentation creation
• Validation of heuristic application
Turned waiting time into preparation time
3. Designed email communication as part of the experience
4. Redesigned Magpie's UI (Game Changer!)
Research & Insights
CRITICAL (4 issues)
Information Overload (Heuristic 8: Aesthetic & Minimalist Design)
Information overload with 11 simultaneous columns creates cognitive paralysis. The interface prioritizes data completeness over decision-making speed, overwhelming users instead of guiding them.
Data Freshness Unclear (Heuristic 1: Visibility of System Status)
Subtle data refreshness indicator with icon - "Last Updated" timestamp missing, Users can't predict the icon is an action button that refreshes data.
Users can't tell if data is current or stale
UX fix for this?
→ Add prominent banner: "✓ Data current as of 2:47 PM today" with refresh indicator
Filter Status Hidden (Heuristic 1: Visibility of System Status)
Movement in metrics "Up: 5, Down: 15" doesn't show the timeframe as to when this movement occurred.
Timeframe was important for the user to make informed decisions
Possible UX Solution:
→ → Make time-bound: "Last 7 days: ↑5 ↓15 →10 Not ranking: 12"
Data Freshness Unclear (Heuristic 1: Visibility of System Status)
Subtle data refreshness indicator with icon - "Last Updated" timestamp missing, Users can't predict the icon is an action button that refreshes data.
Users can't tell if data is current or stale
UX fix for this?
→ Add prominent banner: "✓ Data current as of 2:47 PM today" with refresh indicator
Process
From v0 Prototype to High-Fi Design in Figma
Here’s how the dataset filter feature evolved…
Solution
Streamlining Complex Workflows Through AI-Accelerated Desige :
Element 1 : Contextualization
Before Catalog-First Navigation
Users were for forced to browse through all available catalogs before initiating a search, adding unnecessary navigation steps and cognitive overhead to research workflows.
After Search-first Experience
Replaced an entire catalogs search page with a hero-driven landing page with integrated catalog selection that enables immediate search while maintaining discovery through contextual filtering and progressive disclosure.
Element 2 : Consolidation
Before: Scattered Information Architecture
Users navigated to individual catalog pages to view metadata, then accessed a separate filter panel to search within that catalog creating a disconnect between browsing and searching.
After: Organized Information Sidebar
A structured left sidebar consolidates all vital information into an organized, expandable menu. Users can now access source links, sharing options, metadata, and licensing from a single, predictable location with advanced filters grouped
Element 3 : Scaffolding
Before: Unguided Search Experience
Users faced a blank search box with no guidance on what types of climate data existed. PhD students and climate researchers searched blindly, often using incorrect terminology or missing relevant datasets entirely because they didn't know what categories were available.
After: Category-Based Search Guidance
A clear "Sub-Categories" section displays the five most-used data types with checkboxes and info icons. Users can now filter by data types with descriptions explaining each category. This guides researchers toward the right data type immediately.
Element 4 : Prioritization
Element 6 : Law of Similarity : Leveraging mental models
Before: Custom Temporal Filter Pattern
The original temporal filter used a basic text input with calendar picker, a pattern unfamiliar to researchers. Users could only select simple date ranges with no recurrence support. A use case scenario where complex temporal queries like "precipitation data every month June–December from 1955–1961" required workarounds or multiple searches.
After User-Driven Advanced Filters
The redesigned filter adopts Google Calendar's proven interaction pattern: start/end dates with time inputs, recurrence options ("Does Not Repeat" / "Never" / "Ends On"), and "All Day" checkbox. Users instantly recognize the interface and understand how to configure complex temporal queries without training.
Real Use Case: Monsoon Analysis
Element 6 : Accomodaiton
Before: Spatial Filter with bounding box
Simple checkbox: "Filter by spatial extent" Single interaction method: "Click on the map to add a bounding box"
After: Spatial Filter with user centric selection options.
Spatial Filter with city name input, lat/long input, bounding box selection, location pin and shape file upload.
Element 7 : Contextual Metadata
Reflection
From pain points to power features
89%
Reduction in filter configuration errors
76%
Faster complex filter configuration
8X
Reduction in unnecessary search queries
This project pushed me to navigate multiple constraints while maintaining design quality and user experience.
Reflection
Navigating Ambiguity, Overcoming Challenges
This project pushed me to navigate multiple constraints while maintaining design quality and user experience.
🤖 AI INTEGRATION
Balancing AI efficiency with design quality, managing inconsistent outputs, and development handoff issues.
What I did :
👩🎨 DESIGN’S NEW ROLE ON THE TEAM
💡KEY LEARNINGS
AI as a Design Accelerator
I learned to leverage AI for rapid ideation and prototyping while maintaining design quality through strategic prompt engineering.
Prioritization Under Constraints
Learned to focus on MVP essentials over nice-to-haves, delivering meaningful progress within technical and timeline constraints.
💡CHECK OUT RECENT VERSION OF AI ACCELERATED DESIGN…
More Projects
Strategic UX Redesign
SEO AI
SEO AI helps grow traffic, boost ROI, increase LLM visibility, and cut SEO costs
Year :
2025-2026
Industry :
Digital Marketing SaaS / SEO Technology
Client :
Search Atlas
Project Duration :
2 weeks
Tools:
Figma, Claude.Ai, Figma Make, Nielsen Heuristics
Search Atlas provides enterprise-grade SEO AI OTTO for digital marketers. Their Keyword Rank Tracker is a critical feature used daily by SEO professionals to monitor hundreds of keywords across multiple search engines. However, user feedback indicated frustration with the interface's complexity and difficulty completing routine tasks efficiently.
As a Senior UI/UX designer....
How might I redesign the Keyword Rank Tracker to reduce cognitive load and improve task completion speed for SEO professionals managing large keyword portfolios, while maintaining access to comprehensive data?
Problem
The Challenge
The current dashboard presents 11 simultaneous columns with 15 elements per row, creating cognitive paralysis that overwhelms users instead of guiding them toward informed decisions.
Core Finding
The inter face prioritizes data completeness over usability, requiring a strategic shift toward progressive disclosure
• 10 Critical Issues Identified
• 8 Heuristics Violated • 4 High Priority Fixes
Process
Strategic Approach
I employed a systematic heuristic evaluation framework combined with AI-assisted analysis to maximize efficiency and thoroughness:
2. AI-Assisted Analysis with Claude.ai
This approach demonstrates strategic tool selection — using AI not as a replacement for design expertise, but as a productivity multiplier that allowed me to focus cognitive energy on strategic decisions rather than tactical execution.
Leveraging Claude.ai as a design thinking partner accelerated my analysis process
• Rapid iteration on solution concepts
• Generation of multiple design alternatives
• Documentation and presentation creation
• Validation of heuristic application
Turned waiting time into preparation time
3. Designed email communication as part of the experience
4. Redesigned Magpie's UI (Game Changer!)
Research & Insights
CRITICAL (4 issues)
Information Overload (Heuristic 8: Aesthetic & Minimalist Design)
Information overload with 11 simultaneous columns creates cognitive paralysis. The interface prioritizes data completeness over decision-making speed, overwhelming users instead of guiding them.
Data Freshness Unclear (Heuristic 1: Visibility of System Status)
Subtle data refreshness indicator with icon - "Last Updated" timestamp missing, Users can't predict the icon is an action button that refreshes data.
Users can't tell if data is current or stale
UX fix for this?
→ Add prominent banner: "✓ Data current as of 2:47 PM today" with refresh indicator
Filter Status Hidden (Heuristic 1: Visibility of System Status)
Movement in metrics "Up: 5, Down: 15" doesn't show the timeframe as to when this movement occurred.
Timeframe was important for the user to make informed decisions
Possible UX Solution:
→ → Make time-bound: "Last 7 days: ↑5 ↓15 →10 Not ranking: 12"
Data Freshness Unclear (Heuristic 1: Visibility of System Status)
Subtle data refreshness indicator with icon - "Last Updated" timestamp missing, Users can't predict the icon is an action button that refreshes data.
Users can't tell if data is current or stale
UX fix for this?
→ Add prominent banner: "✓ Data current as of 2:47 PM today" with refresh indicator
Process
From v0 Prototype to High-Fi Design in Figma
Here’s how the dataset filter feature evolved…
Solution
Streamlining Complex Workflows Through AI-Accelerated Desige :
Element 1 : Contextualization
Before Catalog-First Navigation
Users were for forced to browse through all available catalogs before initiating a search, adding unnecessary navigation steps and cognitive overhead to research workflows.
After Search-first Experience
Replaced an entire catalogs search page with a hero-driven landing page with integrated catalog selection that enables immediate search while maintaining discovery through contextual filtering and progressive disclosure.
Element 2 : Consolidation
Before: Scattered Information Architecture
Users navigated to individual catalog pages to view metadata, then accessed a separate filter panel to search within that catalog creating a disconnect between browsing and searching.
After: Organized Information Sidebar
A structured left sidebar consolidates all vital information into an organized, expandable menu. Users can now access source links, sharing options, metadata, and licensing from a single, predictable location with advanced filters grouped
Element 3 : Scaffolding
Before: Unguided Search Experience
Users faced a blank search box with no guidance on what types of climate data existed. PhD students and climate researchers searched blindly, often using incorrect terminology or missing relevant datasets entirely because they didn't know what categories were available.
After: Category-Based Search Guidance
A clear "Sub-Categories" section displays the five most-used data types with checkboxes and info icons. Users can now filter by data types with descriptions explaining each category. This guides researchers toward the right data type immediately.
Element 4 : Prioritization
Element 6 : Law of Similarity : Leveraging mental models
Before: Custom Temporal Filter Pattern
The original temporal filter used a basic text input with calendar picker, a pattern unfamiliar to researchers. Users could only select simple date ranges with no recurrence support. A use case scenario where complex temporal queries like "precipitation data every month June–December from 1955–1961" required workarounds or multiple searches.
After User-Driven Advanced Filters
The redesigned filter adopts Google Calendar's proven interaction pattern: start/end dates with time inputs, recurrence options ("Does Not Repeat" / "Never" / "Ends On"), and "All Day" checkbox. Users instantly recognize the interface and understand how to configure complex temporal queries without training.
Real Use Case: Monsoon Analysis
Element 6 : Accomodaiton
Before: Spatial Filter with bounding box
Simple checkbox: "Filter by spatial extent" Single interaction method: "Click on the map to add a bounding box"
After: Spatial Filter with user centric selection options.
Spatial Filter with city name input, lat/long input, bounding box selection, location pin and shape file upload.
Element 7 : Contextual Metadata
Reflection
From pain points to power features
89%
Reduction in filter configuration errors
76%
Faster complex filter configuration
8X
Reduction in unnecessary search queries
This project pushed me to navigate multiple constraints while maintaining design quality and user experience.
Reflection
Navigating Ambiguity, Overcoming Challenges
This project pushed me to navigate multiple constraints while maintaining design quality and user experience.
🤖 AI INTEGRATION
Balancing AI efficiency with design quality, managing inconsistent outputs, and development handoff issues.
What I did :
👩🎨 DESIGN’S NEW ROLE ON THE TEAM
💡KEY LEARNINGS
AI as a Design Accelerator
I learned to leverage AI for rapid ideation and prototyping while maintaining design quality through strategic prompt engineering.
Prioritization Under Constraints
Learned to focus on MVP essentials over nice-to-haves, delivering meaningful progress within technical and timeline constraints.
💡CHECK OUT RECENT VERSION OF AI ACCELERATED DESIGN…
More Projects
Strategic UX Redesign
SEO AI
SEO AI helps grow traffic, boost ROI, increase LLM visibility, and cut SEO costs
Year :
2025-2026
Industry :
Digital Marketing SaaS / SEO Technology
Client :
Search Atlas
Project Duration :
2 weeks
Tools:
Figma, Claude.Ai, Figma Make, Nielsen Heuristics
Search Atlas provides enterprise-grade SEO AI OTTO for digital marketers. Their Keyword Rank Tracker is a critical feature used daily by SEO professionals to monitor hundreds of keywords across multiple search engines. However, user feedback indicated frustration with the interface's complexity and difficulty completing routine tasks efficiently.
As a Senior UI/UX designer....
How might I redesign the Keyword Rank Tracker to reduce cognitive load and improve task completion speed for SEO professionals managing large keyword portfolios, while maintaining access to comprehensive data?
Problem
The Challenge
The current dashboard presents 11 simultaneous columns with 15 elements per row, creating cognitive paralysis that overwhelms users instead of guiding them toward informed decisions.
Core Finding
The inter face prioritizes data completeness over usability, requiring a strategic shift toward progressive disclosure
• 10 Critical Issues Identified
• 8 Heuristics Violated • 4 High Priority Fixes
Process
Strategic Approach
I employed a systematic heuristic evaluation framework combined with AI-assisted analysis to maximize efficiency and thoroughness:
2. AI-Assisted Analysis with Claude.ai
This approach demonstrates strategic tool selection — using AI not as a replacement for design expertise, but as a productivity multiplier that allowed me to focus cognitive energy on strategic decisions rather than tactical execution.
Leveraging Claude.ai as a design thinking partner accelerated my analysis process
• Rapid iteration on solution concepts
• Generation of multiple design alternatives
• Documentation and presentation creation
• Validation of heuristic application
Turned waiting time into preparation time
3. Designed email communication as part of the experience
4. Redesigned Magpie's UI (Game Changer!)
Research & Insights
CRITICAL (4 issues)
Information Overload (Heuristic 8: Aesthetic & Minimalist Design)
Information overload with 11 simultaneous columns creates cognitive paralysis. The interface prioritizes data completeness over decision-making speed, overwhelming users instead of guiding them.
Data Freshness Unclear (Heuristic 1: Visibility of System Status)
Subtle data refreshness indicator with icon - "Last Updated" timestamp missing, Users can't predict the icon is an action button that refreshes data.
Users can't tell if data is current or stale
UX fix for this?
→ Add prominent banner: "✓ Data current as of 2:47 PM today" with refresh indicator
Filter Status Hidden (Heuristic 1: Visibility of System Status)
Movement in metrics "Up: 5, Down: 15" doesn't show the timeframe as to when this movement occurred.
Timeframe was important for the user to make informed decisions
Possible UX Solution:
→ → Make time-bound: "Last 7 days: ↑5 ↓15 →10 Not ranking: 12"
Data Freshness Unclear (Heuristic 1: Visibility of System Status)
Subtle data refreshness indicator with icon - "Last Updated" timestamp missing, Users can't predict the icon is an action button that refreshes data.
Users can't tell if data is current or stale
UX fix for this?
→ Add prominent banner: "✓ Data current as of 2:47 PM today" with refresh indicator
Process
From v0 Prototype to High-Fi Design in Figma
Here’s how the dataset filter feature evolved…
Solution
Streamlining Complex Workflows Through AI-Accelerated Desige :
Element 1 : Contextualization
Before Catalog-First Navigation
Users were for forced to browse through all available catalogs before initiating a search, adding unnecessary navigation steps and cognitive overhead to research workflows.
After Search-first Experience
Replaced an entire catalogs search page with a hero-driven landing page with integrated catalog selection that enables immediate search while maintaining discovery through contextual filtering and progressive disclosure.
Element 2 : Consolidation
Before: Scattered Information Architecture
Users navigated to individual catalog pages to view metadata, then accessed a separate filter panel to search within that catalog creating a disconnect between browsing and searching.
After: Organized Information Sidebar
A structured left sidebar consolidates all vital information into an organized, expandable menu. Users can now access source links, sharing options, metadata, and licensing from a single, predictable location with advanced filters grouped
Element 3 : Scaffolding
Before: Unguided Search Experience
Users faced a blank search box with no guidance on what types of climate data existed. PhD students and climate researchers searched blindly, often using incorrect terminology or missing relevant datasets entirely because they didn't know what categories were available.
After: Category-Based Search Guidance
A clear "Sub-Categories" section displays the five most-used data types with checkboxes and info icons. Users can now filter by data types with descriptions explaining each category. This guides researchers toward the right data type immediately.
Element 4 : Prioritization
Element 6 : Law of Similarity : Leveraging mental models
Before: Custom Temporal Filter Pattern
The original temporal filter used a basic text input with calendar picker, a pattern unfamiliar to researchers. Users could only select simple date ranges with no recurrence support. A use case scenario where complex temporal queries like "precipitation data every month June–December from 1955–1961" required workarounds or multiple searches.
After User-Driven Advanced Filters
The redesigned filter adopts Google Calendar's proven interaction pattern: start/end dates with time inputs, recurrence options ("Does Not Repeat" / "Never" / "Ends On"), and "All Day" checkbox. Users instantly recognize the interface and understand how to configure complex temporal queries without training.
Real Use Case: Monsoon Analysis
Element 6 : Accomodaiton
Before: Spatial Filter with bounding box
Simple checkbox: "Filter by spatial extent" Single interaction method: "Click on the map to add a bounding box"
After: Spatial Filter with user centric selection options.
Spatial Filter with city name input, lat/long input, bounding box selection, location pin and shape file upload.
Element 7 : Contextual Metadata
Reflection
From pain points to power features
89%
Reduction in filter configuration errors
76%
Faster complex filter configuration
8X
Reduction in unnecessary search queries
This project pushed me to navigate multiple constraints while maintaining design quality and user experience.
Reflection
Navigating Ambiguity, Overcoming Challenges
This project pushed me to navigate multiple constraints while maintaining design quality and user experience.
🤖 AI INTEGRATION
Balancing AI efficiency with design quality, managing inconsistent outputs, and development handoff issues.
What I did :
👩🎨 DESIGN’S NEW ROLE ON THE TEAM
💡KEY LEARNINGS
AI as a Design Accelerator
I learned to leverage AI for rapid ideation and prototyping while maintaining design quality through strategic prompt engineering.
Prioritization Under Constraints
Learned to focus on MVP essentials over nice-to-haves, delivering meaningful progress within technical and timeline constraints.