Dell Technologies Service Portal Employee Experience

IT Service Portal Redesign

Transformed a legacy IT service portal from a ticket-submitting form into an intelligent self-service experience — dramatically deflecting tickets, accelerating resolution, and improving employee satisfaction at enterprise scale.

55%Ticket Deflection
48%Faster Resolution
34%Higher CSAT
28%Lower Cost-per-Ticket

The Problem

  • Business: 65% of IT tickets were password resets, software access, or FAQ-level questions — burning $2.8M/year in agent time.
  • Users: Employees couldn't find answers; the portal's search returned irrelevant results 73% of the time.
  • Constraints: Existing ServiceNow backend could not be replaced; redesign had to work within current workflow engine.

The Solution

  • Strategy: Shift from "submit a ticket" to "find and fix yourself" — intelligent search, guided workflows, and predictive routing.
  • Design: NLP-powered search, conversational request wizard, real-time ticket tracker, and proactive status notifications.
  • Validation: 18 usability sessions; A/B on search relevance with 3,200 employees; 4-week pilot with HR and Sales.

Executive Summary

The company's IT service portal was a legacy form-based system where every employee interaction started with "Submit a ticket." The result was a bloated queue where 65% of tickets were simple, repeatable requests that could have been self-served — costing the organization $2.8M annually in agent time alone.

I led the end-to-end redesign of the portal experience — from search and request submission to tracking and resolution — introducing intelligent self-service, conversational request wizards, and proactive status updates. The outcome: more than half of common tickets were deflected before reaching an agent.

Business Problem

  • 65% of tickets were password resets, software requests, or FAQ-level — all solvable without an agent
  • Average ticket resolution time of 4.2 days, with 30% of tickets reopening due to incomplete initial info
  • $2.8M annual cost in agent time handling deflectable requests
  • Portal search returned irrelevant results 73% of the time — pushing users straight to ticket submission
  • Employee NPS for IT support was -12, driven by wait times and poor communication

User Research

Conducted 18 contextual interviews and 400+ ticket categorization reviews across HR, Sales, Engineering, and Finance. Also analyzed 6 months of portal search logs and abandonment funnels.

Insight 01

Employees didn't want to file tickets — they wanted instant answers. "I only submit because I can't find it."

Insight 02

73% of portal searches were 1–3 keywords; the legacy search engine only matched exact article titles.

Insight 03

30% of reopened tickets lacked required info — the form had 18 fields with no guidance on what mattered.

Insight 04

Employees checked ticket status 4.3 times on average — but status updates were only visible inside the portal.

Discovery Insights

  • A conversational request wizard that asked 4 smart questions could replace an 18-field form
  • NLP-powered semantic search would deflect 40%+ of FAQ-level queries immediately
  • Proactive email/Slack status updates would eliminate 70% of status-check visits
  • Predictive routing — auto-categorizing and suggesting solutions — would reduce misrouted tickets by 25%

Journey Mapping

Mapped the employee support journey from problem awareness through resolution. Identified the "frustration cliff" — the moment employees gave up on self-service and submitted a ticket. The primary cause was search failure followed by an overwhelming form. Redesigning search and simplifying submission were the highest-impact fixes.

Product Strategy

Defined a self-service-first strategy with three pillars:

  • Intelligent search: NLP-powered semantic search across knowledge base, policies, and common fixes
  • Conversational request wizard: Replace the 18-field form with a 4-question guided flow that auto-categorizes and suggests fixes
  • Proactive communication: Push status updates via email and Slack; reduce portal check-ins

Success metrics: ticket deflection rate, average resolution time, cost-per-ticket, and employee CSAT.

Design Exploration

Explored 3 directions for the new portal: a traditional knowledge-base homepage, an AI chatbot-first experience, and a hybrid search-and-guide approach. The hybrid won in testing — it served both explorers (search) and novices (guided workflows) without forcing either into an unfamiliar pattern.

IT Service Portal — Semantic Search & Self-Service

Designed the request wizard to feel like a conversation while auto-populating backend ServiceNow fields invisibly — improving UX without requiring a platform migration.

Validation

  • 18 usability sessions with employees across 4 departments
  • A/B test on search relevance with 3,200 employees over 4 weeks
  • 4-week pilot with HR and Sales teams — measuring deflection and resolution time
  • Accessibility audit (WCAG 2.1 AA) and mobile responsiveness testing

Final Solution

Shipped an intelligent IT service portal with semantic search, conversational request wizard, real-time ticket tracking, proactive status notifications, and predictive auto-routing — all integrated with the existing ServiceNow backend.

Business Results

55%Of common tickets deflected via self-service
48%Faster average ticket resolution time
34%Improvement in employee CSAT score