top of page
OPTIMISED FIELD PLANNING
From 25 Hours to 25 Minutes: Redesigning Territory Planning with AI
Project Overview
My Role
Lead UI/UX Designer
Team
4 members
Timeline
12 weeks
Tools
Figma, Figjam, Excel
.png)
IMPACT
75%
Review Time Reduction
12-18%
Travel Time Reduction
20%
Coverage Consistency Improvement
THE PROBLEM
What We Discovered
Field Level Managers were spending 25 hours every month manually balancing sales territories. They would download reports, open spreadsheets, and move zip codes around one by one, hoping their changes would reduce travel time and balance workloads.
The Real Pain point
⏰
Too Much Time, Too Little Insight
Managers spent entire days editing zip codes. By the time they finished, market conditions had already changed.
👾
Playing Territory Tetris
Moving one zip code affected everything else. A change that fixed one rep's workload often created problems for another. There was no way to see the ripple effects.
📊
Guessing, Not Knowing
Will this reduce travel time or increase it? Will my reps be upset? Am I breaking up important customer relationships? These questions kept managers up at night.
🔂
The Same Problems, Every Month
Without continuous monitoring, issues piled up. By the time the monthly report came out, territories were already badly misaligned.
The Business Impact
This wasn't just frustrating for managers : it was costing the company money
-
Sales reps spent 12+ hours per week just driving between appointments
-
Inconsistent territory coverage meant some high-value customers weren't getting enough attention
-
Misaligned workloads led to burnout for some reps while others had too little to do
-
When territories changed, customer relationships suffered because handoffs were poorly planned
RESEARCH & DISCOVERY
How We Got Started
We spent 3 weeks talking to the people who lived with this problem every day
Who We Talked To
-
8 Field Level Managers across different regions
-
3 Sales Operations Leads
Key Research Activities
🗺️
Journey Mapping Workshop
We brought Field Level Managers together and mapped out their entire monthly process. We counted 29 manual steps just to complete one territory alignment cycle. The room went quiet when we realized how broken the process was.
😖
Pain Point Interviews
We asked everyone: "What's the worst part of your job?" Territory planning came up again and again.
What We Learned
From Field Level Managers
-
"I spend more time in Excel than helping my team sell."
-
"I make changes and just pray they work out."
-
"Every month feels like starting from scratch."
From Sales Operations Leads
-
"Every manager does this differently. There's no consistency."
-
"By the time we catch territory problems, we've already lost deals."
-
"We have all this data but no way to use it proactively."
MEET OUR USERS
Sarah Martinez
Field Level Manager
12 years experience
Oversees 5 Region
Manages 60+ sales reps
Goals
Balance workloads across sales representatives
Minimize travel time and maximize face-to-face customer interactions
Preserve and strengthen customer relationships during territory changes
Make confident decisions backed by data, not guesswork
Frustrations
Spending 25 hours per month on manual territory edits
No way to predict if changes will help or hurt
Finding errors after changes are already implemented
Always reacting to problems instead of preventing them
4 Hours
Current review time per cycle
12 Hours
Avg. weekly travel time per rep
68%
Current coverage consistency score
Michael Chen
Sales Operations Lead
9 years experience
Northeast Division
Oversees 40+ Sales rep
Goals
Create consistency across all regions
Catch territory problems before they impact quarterly targets
Reduce the operational overhead of territory management
Make sure all territory changes follow company policies
Frustrations
Every manager has their own "system" (Excel spreadsheets, gut feelings, etc.)
No visibility into territory changes until after they happen
Spending 45% of his time coordinating territory issues
8% error rate in territory alignments causing downstream problems
8%
Current error rate in alignment
2 Regions
Under Management oversight
45%
Time spent on manual coordination
THE PROCESS : HOW WE GOT THE SOLUTION
Phase 1 : Mapping the Current Workflow (Week 1-2)
We started by understanding exactly how territory planning worked today, mapping every single step of the monthly territory planning process.
Download monthly performance report
(30 minutes)
Review rep-by-rep metrics in Excel
(1 hour)
Identify problem territories manually (2 hours)
Research which zip codes to move
(3 hours)
Calculate potential travel time impacts (4 hours)
Check for territory contiguity issues on a map (2 hours)
Create draft territory changes in Excel (3 hours)
Manually verify each change won't break rules (3hours)
Submit changes and wait 2 weeks for implementation
The Insight
Most of these steps involved hunting for information that already existed in our systems. The computer could do the analysis—humans should make the decisions.
Phase 2: Exploring Solutions (Week 3-4)
Two Concepts We Considered
🤖
Concept A: Fully Automated System
AI makes all territory changes automatically, managers just approve or reject the entire batch.
❌ Why We Rejected It
Managers didn't trust "black box" AI. They wanted to understand the reasoning and have granular control.
😎
Concept B: Smart Assistant
AI continuously monitors territories and alerts managers to problems with specific suggestions they can accept, reject, or modify.
✅ Why We Chose It
Keeps humans in control while eliminating tedious analysis work. Managers get to make the final call.
Phase 3: Designing New Flows (Week 5-6)
We needed to restructure how information was organized and presented.
User receives intelligent
alert notification
Click alert to open detailed suggestion
Navigate through recommendation options
Accept/ reject suggestion
Review and confirm changes
Key Decisions
🔔 Alert-Based System
Instead of making managers hunt for problems in reports, we bring problems to them through intelligent alerts.
📈 Ranked Suggestions
AI analyzes hundreds of possible territory changes and shows only the top recommendations with the biggest impact.
⚖️ Impact Visualization
Before accepting any change, managers see exactly what will happen: workload shifts, travel time changes, customer disruptions.
✋🏻 Guardrails First
AI suggestions automatically respect territory contiguity, workload limits, etc. so managers never see invalid options.
Phase 4: Prototyping & Testing (Week 7-9)
.png)
Round 1 Feedback
✅ "I love that it tells me WHY this is a problem"
❌ "I need to see all my options, not just the top suggestion."
❌ "What if I disagree with the AI? Can I edit this?"
Changes Made
✅ Added ability to view multiple ranked suggestions
✅ Made the AI reasoning more transparent
⚠️"Edit" functionality was descoped
Round 2 Feedback
✅ "This is incredible. I can see the impact before I make the change."
✅ "The travel time visualization makes so much sense."
Further Improvements
✅ Added comparison view showing current vs. proposed territories
Human-in-Loop Design
Accept/reject/edit functionality for all suggestions, with guardrails always applied before display to ensure compliance with business rules
ACCESSIBILITY CONSIDERATIONS
Table View : alternative to map interface for users who cannot interact with spatial visualizations
WCAG AA compliant : contrast ratios for all map overlays and metric tables
Full keyboard navigation : support through alerts and suggestions
PROCESS TRANSFORMATION
Data Review
Download monthly report manually
Continuous monitoring with intelligent alerts
Manual filtering and zip code editing
One-click view of ranked suggestions
Analysis
Manual travel time calculations
Instant metric comparison with disruption/travel impact
Impact Assessment
Implementation
Submit changes and wait for updates
Accept/reject with instant draft updates
Process Steps
Before AI
After AI
QUANTITATIVE RESULTS
Review Time
4 hours
0.42 hours
12 hours
10 hours
Travel Time/Week
-75%
-12-18%
68%
82%
Coverage Consistency
+20%
Error Rate
8%
2%
-75%
Metric
Before AI
After AI
Impact
What This Means
-
Field Level Managers saved 3.75 hours per cycle → reinvested in coaching their teams
-
Sales reps saved 1.5-2 hours per week → 7-10 more customer meetings per month
-
Better coverage consistency → higher customer satisfaction and more strategic account management
-
Fewer errors → reduced frustration for sales reps and fewer support tickets for Sales Ops
WHAT I LEARNED
🎨 Design Lessons
-
Trust is Earned, Not Assumed
Early prototypes had AI making automatic changes for "low-risk" issues. Users hated it. Even when AI was 95% accurate, they wanted to review every change. We learned: give people control, not just automation.
-
Show Your Work
"The system suggests moving these 5 zip codes" wasn't enough. Users needed to know WHY. We added reasoning for every suggestion, and trust went way up.
-
Process Matters as Much as Interface
A beautiful interface on top of a broken workflow is still broken. We had to redesign the entire territory planning process, not just make it look prettier.
🤝 Team & Collaboration
-
What Worked
-
Weekly design reviews with Field Level Managers kept us grounded in reality
-
Pairing with the data scientist helped me understand what AI could and couldn't do
-
Building quick prototypes (even paper) prevented us from investing in wrong ideas
-
Celebrating small wins kept team morale high during the long 12-week project
2. What Was Hard
-
Balancing the needs of two very different user groups (Field Level Managers vs. Sales Ops Leads)
-
Managing stakeholder expectations when we had to push back on requested features
-
Staying focused on the core problem when everyone had ideas for "cool" features
The BOTTOM LINE
We transformed territory planning from a 25-hour monthly burden into a 25-minute strategic exercise. But more importantly, we gave Field Level Managers their time back so they could focus on what matters:
coaching their teams and building relationships with customers.
This wasn't just a design project. It was about making people's work lives better.
bottom of page