Hospital Page (2026)

Using experimentation, analytics, and behavioural insights to improve conversion and user experience

Overview

This project focused on redesigning a high-traffic hospital landing page at Nuffield Health — often the first step in a patient’s private healthcare journey.

The page needed to balance:

  • Trust-building

  • Medical reassurance

  • Information discovery

  • Commercial conversion

Previous A/B tests, surveys, and analytics showed the page was too conversion-focused, and not aligned with how patients actually make decisions.

The opportunity was to transform the page from a single-path funnel into a multi-path decision-support experience.

The Problem

The original page:

  • Pushed users toward one primary conversion goal

  • Buried Nuffield’s strongest USPs & credibility signals

  • Offered limited onward journeys

  • Under-served users in research or consideration mode

Result: Users weren’t rejecting the brand — they were uncertain about next steps or not ready to commit yet.

The Hypothesis

If we redesign the page using behavioural data, clearer messaging, stronger unique selling point (USP) surfacing, and more flexible onward journeys, users will find the right information faster, feel more confident, and show stronger engagement and conversion.

The goal wasn’t just more conversions — but better decision quality.

Research & Data Inputs

We grounded decisions in multiple insight streams:

Quantitative

  • Adobe Analytics conversion & drop-off data

  • Funnel progression & scroll depth

  • Downstream journey tracking

Qualitative

  • On-page poll responses

  • Patient survey feedback

  • Previous A/B test learnings

Behavioural

  • Microsoft Clarity heatmaps & session replays

  • Hesitation, rage-clicks, and abandonment signals

Key Insight

Users landing on this page fall into different intent stages:

  • Ready to book

  • Comparing providers

  • Seeking reassurance

  • Still understanding their condition

A single-path experience could not serve all of them effectively.

Design Process

Wireframing & Concept Exploration

I created four wireframe concepts:

  • 3 concepts aligned to the existing Design System & CMS constraints

  • 1 blue-sky concept to explore future opportunities

This ensured we balanced real-world delivery feasibility with innovation.

Cross-Functional Collaboration

Worked closely with:

  • UX & Content

  • SEO

  • Product

  • Engineering

Together, we shortlisted concepts, built them in CMS, and refined into test-ready UI variants.

Experimentation & UI Variants

Variant A — Conversion-Forward (With Form)

Designed for:

  • High-intent users

  • Faster enquiry completion

  • Shorter decision cycles

Variant B — Exploration-Forward (No Form)

Designed for:

  • Research-mode users

  • Reassurance & education

  • Reduced friction

  • Stronger onward journey discovery

Additional Variables Tested

  • FAQ inclusion to reduce uncertainty

  • USP prominence & framing

  • Location content placement (higher vs lower)

  • Onward journey link hierarchy

A/B Test Setup

  • Control vs Variant testing

  • Minimum test duration: 1 month

  • High traffic ensured statistical confidence

  • Performance tracked across:

    • Conversion rate

    • Engagement & interaction

    • Scroll depth

    • Next-page journey behaviour

Early results showed a short-term dip, but rather than reacting prematurely, we extended the test by 4 weeks — prioritising data integrity over gut reaction.

Results & Performance Impact

Hospital Redesign A-B Test

Engagement & Behaviour Uplift

The redesigned Variant demonstrated:

  • Increased engagement with onward journeys

  • Higher interaction with USP & trust content

  • Improved scroll depth, indicating stronger content relevance

  • Reduced bounce rate, showing better alignment with user intent

  • Stronger next-page progression, especially toward hospital and treatment flows

Journey Flow Improvements

Users were more likely to:

  • Continue into relevant hospital & treatment journeys

  • Explore supporting informational content

  • Follow a considered decision path, rather than abandoning early

This indicated better alignment with patient decision behaviour, not just short-term conversion pressure.

Key Learnings

  • Balanced conversion + education outperforms hard-sell layouts

  • Surfacing USPs & reassurance increases trust & engagement

  • Supporting research behaviour improves downstream conversion

  • Healthcare users value clarity, autonomy, and confidence

  • Optimising for decision quality creates stronger long-term value than forcing urgency

Business & UX Impact

  • Improved first-impression experience for prospective patients

  • Supported both high-intent converters and early-stage researchers

  • Increased engagement with high-value onward journeys

  • Created a repeatable CRO & experimentation model for hospital pages

  • Shifted internal thinking from “push conversion” to “support confident decision-making”

How This Reflects My UX Approach

  • I design for real user intent, not vanity metrics

  • I use data as a decision engine, not decoration

  • I balance commercial performance with ethical, patient-centred UX

  • I’m comfortable working across UX, SEO, Product, and Engineering

  • I treat experimentation as a continuous learning system, not a one-off test

What I’d Explore Next

  • Personalised journeys by condition, urgency, or intent

  • Dynamic trust signals (reviews, clinician credibility, outcomes)

  • Pricing transparency & affordability experiments

  • Segmentation-based CRO for hospital services

  • A rolling hospital optimisation roadmap