Index / Selected Work / Case 01 · Iris
Conversational AI UX Design UX Writing Data Visualisation

Iris, an AI analyst for corporate travel.

Designing a conversational interface that turns complex travel data into instant, trustworthy insights, without adding another dashboard.

Try Iris, live
Plate 01 · Iris answering a high-volume booking query ITILITE · 2024
My Role
Lead Product
DesignerConversational AI · Data viz
Timeline
8 weeksCross-functional with PMs, eng & data
Users
Client admins,
finance & travelEnterprise corporate travel teams
Impact
−50% report
requests / mo.Hours of work → seconds of answer
01 · Context

Travel teams were drowning in data,
starved for answers.

ITILITE helps businesses book, manage and optimise employee travel. Corporate travel teams own a sprawling, opinionated dataset, yet getting to a single answer required exporting spreadsheets, building dashboards manually, or asking an internal analyst to play archaeologist.

The problem wasn't missing data. It was the distance between the question and the answer.

  • Export, pivot, hand-analyse, the default workflow.
  • Dashboard sprawl: too many surfaces, too little signal.
  • Support tickets to CSMs for ad-hoc reporting.
I wish someone could just tell me who's landing when, without manually scanning bookings.
Travel manager, organising a 400-person offsite
Plate 02 · Iris embedded inside the existing insights surface 21 : 9 · Editorial composite
02 · Design challenge

How might travel admins get direct answers from travel data, instantly?

The temptation was to ship another analytics page. We resisted.

Instead, I designed Iris as an AI-first analytics layer that lets admins self-serve insights in natural language, woven into the existing reporting and insights workflows rather than sitting alongside them as a new surface.

The work focused on three things: progressive guidance, clear guardrails, and trust-building patterns, so users could explore data confidently, and the system could decline confidently when it couldn't.

The design decisions.

Four moves that shaped how Iris feels, less an AI feature, more a quiet member of the team.

01

Meaningful prompts

  • Identified questions not already solvable via existing dashboards or reports
  • Partnered with Sales and CSMs to uncover recurring customer gaps
  • Used prompts to demonstrate capability and guide first-time usage
02

Navigation restraint

  • Avoided adding another analytics page to an already crowded system
  • Integrated Iris within existing insights and reporting flows
  • Reduced fragmentation while improving discoverability
03

Edge & error states

  • Designed non-happy paths: thinking, limits, failures, expired queries
  • Wrote UX copy for unsupported or out-of-scope data requests
  • Ensured clarity and trust even when the system couldn't respond
04

Visual consistency

  • AI-generated charts & tables reuse existing patterns, palette, hierarchies
  • Avoided new visual languages, outputs feel reliable, familiar, core
  • Trust and readability over novelty, especially for data-heavy views

The impact.

03 · Outcomes
−50%
Reduction in client report requests, admins self-serve via natural-language queries.
Hours → sec.
Static dashboards replaced by interactive exploration. Spreadsheets compressed to a sentence.
+62%
Discovery uplift after embedding Iris inside existing high-intent touchpoints.
4 outlets
Press coverage at launch, Yahoo Finance, BTN, Travel & Tour World, ITILITE.
Plate 03 · Prompt library
Plate 04 · Chart consistency

Field notes & learnings.

04 · Reflection
i.

AI experiences need strong guardrails.

Designing clear boundaries, error states, and fallback copy was as important as the happy path. Trust is earned in the failures.

ii.

Examples shape behaviour.

The questions shown to users significantly influenced how they explored, and ultimately trusted, the system.

iii.

Integration beats novelty.

AI features gain adoption faster when they feel like part of an existing workflow, not a separate product on the side.

iv.

Designers must think system-wide.

Decisions around navigation, limits, and data scope had more impact than visual polish alone. AI is a system problem.

05 · Final takeaway

Reframing how users reason about their data.

Designing Iris was less about adding AI to the product and more about reframing how users access and reason about their data.

The work required balancing innovation with trust, flexibility with constraints, and discovery with simplicity, treating AI as a system design problem, not just a UI feature.

Press coverage.

06 · Annexe
ITILITE Meet Iris, your AI travel analyst Read ↗ Yahoo Finance ITILITE launches Iris: AI-powered travel analyst Read ↗ Business Travel News ITILITE launches AI analytics tool, Iris Read ↗ Travel & Tour World ITILITE unveils new Iris AI to enhance travel data insights Read ↗
Next chapter · 02 / 05
ITILITE homepage refresh
Open to opportunities

Like what you see?
Let's connect.

Status
Available
Q3 2026
Connect
LinkedIn
Loc.
Bangalore,
India