Eliminating Platform Fragmentation with an AI Chat Interface to Enhance Sales Processes
IntelliAssist streamlines access to information, eliminating the need to search across multiple platforms, which boosts productivity and improves the overall sales process
10,000+
Users
6+
Services in one Interface
85
SUS Score
Details, such as names, workflows, and interfaces, have been modified to maintain confidentiality
Context
Sales reps waste valuable time searching for information across multiple platforms. IntelliAssist, our AI-powered chatbot, solves this by providing instant answers to their questions, boosting productivity and streamlining the sales process.
Business Goals
🚀 Improve sales productivity by providing quick access to relevant information.
💬 Enhance the user experience through a conversational interface and streamlined workflows
🔄 Integrate with existing systems for seamless access to order details and customer data
🔎 Provide instant, accurate answers to users about Dell process management
Users Share Their Experiences to Shape IntelliAssist
We conducted in-depth interviews with five Dell employees representing a diverse range of roles, including sales representatives, data analysts, and project managers. The interviews aimed to understand their daily workflows, challenges, and pain points related to accessing information and completing tasks efficiently
Initial Design
Research-Driven Design
Iterative Prototyping
Data-Informed Refinement
Choosing Interface Styles
Clarity Over Hidden Complexity
I converged on option B after balancing user feedback, brand guidelines, and technical constraints
Employees needed quick answers, not more clicks. Design 2 put sample questions and tools front and center, cutting the guesswork of how to use the AI. Tests showed users adopted it faster, and the familiar Dell branding built trust, which is essential for a company-wide tool
Option A
Familiar First Impression (other chat bots)
Clear Visual Hierarchy
Lower discoverability of sample questions (click to reveal)
Higher Interaction Cost
Option B
Instant discoverability of sample questions
Brand consistency with Dell signature colors
Dense layout
Poor reading pattern
🚨 Balancing Business "wants" with User Guidance
The business loved the yellow tile for sample questions—so we kept it! Turns out, that pop of color isn’t just fun; it guides attention to key actions without breaking trust in Dell’s brand. A small compromise for big clarity
Chat Interface


Streamlined Responses
Prioritizing Clarity & Scannability for Order Details


Verified Answers with Source Transparency
Evaluation Metrics
User Satisfaction (Early Adoption)
System Usability Scale
Sentiment analysis
The pilot group rated the assistant 78/100 — praised "intuitive guidance" and reduced first-time anxiety
Preliminary data from initial launch phase
Workflow Efficiency Gains
Task log analysis
SME interviews
82% task completion rate for core workflows (e.g., order details), saving users ~3.2 minutes per task vs. manual searches
Sampled 2 weeks of post-launch data.
Self-Service Momentum
Escalation rate tracking
58% of inquiries resolved without human support; repeat users hit 72% self-resolution.
Early indicator of learnability and trust.
Increased Decision Confidence
Pre/post-launch surveys
Behavioral telemetry
68% of users reported higher confidence in task outcomes when using IntelliAssist
Measured across 18 observed workflows
Navigating Real-World Constraints
⏳Tight Deadlines for Impact
Accelerated timelines prioritized MVP delivery over exploratory research
Focused on high-impact, SME-backed workflows (e.g., order status checks) to align with stakeholder goals
🔒Resource & Access Limitations
Restricted access to cross-departmental users and historical data due to compliance policies
Leveraged sales-team SMEs as proxies for edge cases and validated designs against existing analytics
🌫️Ambiguous Requirements
Early-stage AI capabilities lacked clear technical boundaries, creating design ambiguities
Adopted iterative prototyping to align engineering feasibility with user needs