How Did an AI-Ready Modular System Fix a Luxury Travel Identity Crisis?

• My Role: Lead Product Designer (Strategy, UI/UX, Logic Architecture)
• Team: CMO, Head of Sales, SEO Lead, Frontend Engineers
• Timeline: 3 Months (Discovery ➔ Release 1.0)
• Core Goal: Realign the desktop experience with the luxury service model & scale SEO operations.

1. The Challenge: When a Concierge Wears a Search Engine Mask

The Business Reality

Business-Class.com is a premium human-powered service, not an automated algorithm.

The UX Disconnect

The legacy homepage mimicked a standard search engine. Users entered dates expecting instant results but were met with a lead form. This "Expectation Mismatch" caused frustration and high bounce rates. Users treated the site like an automated aggregator, entering with a "DIY" mindset.

The Strategic Pivot

Instead of fighting the user's mental model ("I want to search"), I adapted the interface to guide them gently toward the service model ("I need an expert"), establishing immediate trust for high-ticket purchases ($5k+).

2. Information Gain: The "Recent Searches" Fallacy

The Failed Hypothesis

I assumed that business class travelers have a non-linear decision journey, so a "Recent Searches" shortcut would be critical for returning users.

The Reality & Pivot

Post-launch data showed unexpectedly low engagement. Instead of helping, it competed for attention with the "Popular Destinations" list, adding cognitive load. Following the data, we decided to deprecate this module in the next sprint to declutter the interface. This reinforced that reducing decision fatigue is more important than offering theoretical convenience features.

3. The Search Widget: Reducing Cognitive Barriers

To fix the mismatch, I redesigned the Search Widget to act as an Assistant, lowering the Interaction Cost to near zero.

Semantic Shift

Leveraging data from our mobile wins, I reframed the primary CTA. I changed the copy from "Search Flights" to "Check Your Price". The Logic: "Search" implies work (browsing), while "Check Price" implies utility (getting an answer). This subtle semantic shift reduced the perceived commitment needed to start the funnel.

Intelligent Assistance (The "Wizard of Oz")

• Popular Destinations: I introduced a quick-select list in the dropdown. We adopted a "Wizard of Oz" strategy—launching with a manually curated list (MVP) to validate user engagement before committing engineering resources to build a dynamic sorting algorithm.
• Smart Defaults: Implemented IP-based detection to pre-fill "Origin" and "Dates," solving "Empty State Anxiety."

The "Fast Lane"

I carved out a dedicated visual space for a "Call for Exclusive Discount" CTA inside the widget, routing urgent buyers directly to agents.

4. Technical Trade-off: The Video Background

The Conflict

Marketing stakeholders pushed for a video background to enhance the "Luxury Feel."

My Approach & Compromise

While video often hurts performance, I compromised by optimizing the asset and implementing a static "skeleton" image for instant loading. This balanced the aesthetic goals with strict Core Web Vitals (LCP) requirements.

5. Replacing Vague Marketing with Tangible Data

The Problem

The legacy "Best Deals" section displayed generic destinations (e.g., "London from $1000") that lacked contextual relevance for the user.

The Transformation

I renamed the section to "Recent Client Savings" and changed the focus from Advertising to Evidence.

• Data-Driven Relevance: We analyzed sales data to identify the 300 top routes that generate 70% of revenue.
• Anchor Pricing: Instead of a flat price, I introduced a comparison: "Retail Price" vs "Our Price", highlighting the specific savings amount (e.g., "Saved $2,549").
• Personalization: Deals are now filtered by the user's IP location (e.g., showing flights departing from their actual city).

6. Balancing Aggressive Sales Goals with User Focus

The Conflict

The Sales Team pushed for an aggressive, auto-opening chat to maximize lead volume.

The Defense

I demonstrated that interrupting high-intent search behaviors with low-intent chat prompts was diluting our lead quality. We agreed to shift the strategy from "Aggressive Acquisition" to "Contextual Assistance."

The Solution: "Smart Trigger" Protocol

I engineered a context-aware choreography for the chat modal:

• Respect Focus: Chat is suppressed while the user types in the Search Widget.
• Idle Logic: Chat only engages after 10–15s of inactivity (Teaser) or 35s (Full Open).
• Conflict Resolution: If Chat opens, the Hero Carousel pauses immediately to prevent visual overload.

7. Empowering Marketing with AI-Ready Modular Design

The Bottleneck

Marketing needed to launch 300+ route-specific SEO pages but was limited by development resources.

The Fix

I architected a modular Design System of 15+ flexible blocks (Hero, Reviews, FAQ, Trust) integrated directly into the CMS.

Governance & AI

I designed the modules with strict constraints (baked-in padding, typography rules) to ensure visual integrity remains high, even when Marketing builds pages autonomously. The system now leverages AI to populate SEO-optimized text, reducing page production time by 95%.

8. The Verdict: Higher Intent & Faster Operations

Outcomes:

• +27% Increase in Click-through to Booking.
• Operational Velocity: Massive reduction in Time-to-Market for SEO pages.
• Accessibility: Achieved WCAG AA compliance by rebalancing contrast and typography.