Embedded Files
Exploring inspiration-led discovery to create intuitive, engaging and cross-category shopping experiences – Driving £17.5M in revenue
Overview
John Lewis, a UK retailer known for quality and service, saw an opportunity to evolve its search experience. While effective at helping users find products it lacked discovery.
With competitors integrating inspiration-led search inspired by social media behaviours, John Lewis aimed to transform it’s search experience from a functional tool into an engaging discovery journey.
The Problem
Functional over inspirational:
Search is used by 35% of website traffic, but 60% of customers shop within a single department, limiting product discovery.Competitive Gap:
Competitors integrate inspiration-driven search experiences, leveraging social media inspired behaviours
The Challenge
How might we inspire customers through search without disrupting findability?
How can we encourage cross-category exploration and discovery through search?
Define how inspiration fits into user search behaviours
The Results
Defined two target audiences: “Browsers Disguised as Searchers” and “Busy Inspiration Seekers"
Developed and tested six concepts enhancing search-driven inspiration
Delivered four MVP ready recommendations for product teams to test at scale
Integrated concepts into product team roadmaps for future iterations
A/B testing on app for one concept led to +3.6% revenue per session uplift and a projected £17.5M full year benefit
Team & My role
Collaborated within a cross-functional squad of four, including a UX Designer, Product Manager and User Researcher to explore new search experiences. Selected as the first UI lead for an R&D project, leveraging my cross-team experience and stakeholder relationships.
Design and research execution:
Led UI design, ensuring brand and design system alignment
Co-led end-to-end design process following a user-centred approach
Audited and analysed existing John Lewis journeys across web and app to identify gaps
Conducted competitor analysis across 20+ retail and non-retail brands to identify emerging search trends
Led the design and prototyped 3 of the 6 concepts using Figma, iterating based on research insights
Supported user testing, contributed to test plans and shaped insights that informed final concepts
Collaboration and stakeholder engagement
Led weekly stakeholder updates, keeping teams informed
Facilitated two ideation workshops with product and design teams
Presented findings and recommendations to stakeholders
Integrated concepts into product roadmaps through collaboration
8
Week project
4
Team members
2
Ideation workshops
20
User interviews
Approach & Process
Faced with an open-ended brief and differing stakeholder expectations, Sprint Zero was key to aligning our new team. We quickly established a shared understanding, defining goals, scope, timelines and responsibilities using a user-centred approach.
By combining strategy, research and design execution, we ensured our concepts weren’t just visually engaging but also based on user needs and behaviours, helping to redefine how search can inspire and guide.
Key stages
Discovery – Gathered insights, conducted competitor analysis and analysed user research to understand user mindsets
Ideation – Translated insights into ‘How might we’ statements, co-facilitated workshops and prioritised ideas
Design & Prototyping – Created low-fidelity concepts, developed high-fidelity prototypes in Figma and refined designs into testable concepts
Testing & Validation – Conducted usability testing, synthesised insights, presented recommendations based on user needs and business goals
The double diamond framework illustrates the project timeline, highlighting key phases and activities carried out at each stage to ensure our focus remained on user needs
1. Research & Discovery – Understanding Users Search Behaviours & Mindsets
Through three weeks of in-depth research we explored the existing landscape by:
Reviewing existing insights & data – Analysed internal and external reports and data to identify common search journeys and user behaviours
Mapped existing John Lewis search experience – Documented customer journeys across web and app using Miro to evaluate how search currently supports discovery and cross-sell opportunities
Stakeholder & product team alignment – Conducted kickoff sessions to align goals and gather insights
Competitor analysis – Assessed 20 retail and non-retail examples to draw inspiration beyond retail and identify emerging user behaviours across other industries
User interviews (10 Sessions) – Interviewed John Lewis and non-John Lewis customers to identify inspiration-driven shopping behaviours
Competitor examples
Pull & Bear App
Anticipates user needs before searching
M&S App
Provides recommended products and alternative methods to search through voice and barcode scanning
Pinterest
App
Shows content relating to search
YouTube
Keeps users engaged with the result whilst allowing users to scroll through other content
Competitor analysis audit to identify key themes and opportunities
Used Content Square to analyse data and user behaviour across existing customer journeys
During discovery we learnt that:
Summary of discovery insights and questions we needed to define through the next phases of the project.
2. Define & Ideate – Refining Scope And Turning Insights Into Concepts
How far could search evolve without sacrificing usability?
What we now know:
Key users:
“Browsers disguised as searchers” – Enter search without clear intent, often use brand or broad queries, expect efficient results
"Busy Inspiration Seekers" – Appreciate inspiration but have limited time, require quick, engaging ways to explore
Inspiration’s role in search:
A behaviour that encourages continued exploration, resulting in a positive, satisfying outcome
Builds confidence in purchases and encourages discovery of additional items
Search journey gaps:
Lack of support for users with weak or undefined search intent
Visually inconsistent or unappealing results fails to engage users
No features to encourage continued exploration beyond results
Defined approach:
Refining scope:
Mobile app – App users are more likely to search than browse
Home & Fashion categories
Visually led and ideal for delivering inspiration
Fashion is the best performing category for ‘get to product’ and conversion rate
Turning insights into concepts:
Framing the Problem – Created seven “How might we” statements
Collaborative Ideation – Co-led two workshops with cross-functional teams to turn insights into solutions
Prioritised concepts – Used dot voting in ideation workshops to prioritise concepts based on user value and feasibility
Seven 'How might we' statements generated through discovery insights covering the whole search journey
Common home and fashion customer search journeys used to set the scene for ideation sessions
Top ideas voted on by ideation workshop attendees, based on user wants and needs
3. Design & Testing – Evolving Concepts & Validating Through User Feedback
With just over a week to refine ideas, the UX designer and I divided tasks to maximise efficiency.
From ideas to testable designs:
Mapped full search journeys – before, during and after search
Created user flows – Defined interactions and decision points
Designed low-fidelity mock-ups – Visualised early-stage solutions, translated prioritised ideas into structured design concepts
Built high-fidelity prototypes – Developed six test-ready concepts in Figma
Refinement and testing:
Rapidly iterated based on insights – Regularly reviewed ideas against user needs
Identified two key user missions – ‘Handheld’ for guided inspiration and ‘Mission’ for goal orientated searches
Consolidated smaller features – Combined ideas into larger concepts to create more impactful experiences
Led end-to-end design – for three of the six concepts (1, 2, and 5)
Collaborated on usability testing – Observed sessions, synthesised findings and collated recommendations based on feedback
Ideation distillation for concepts using opportunity matrix scoring to prioritise approaches
An opportunity matrix was established to score concepts based on our expertise and learnings developed over the 8 week project.
User flow offering a ‘handholding’ journey with expert support and personalised recommendations for customers who need help finding the right product.
Phase 1 - Concept Exploration – Re-thinking search experiences
With limited time, we sketched ideas and leveraged design system components to quickly mock-up and rapidly iterate. Each concept was aligned to customer wants and needs, ensuring they were relevant, engaging and functional.
Idea 1 – Inspired search via tab (Handheld search)
Idea 2 – Image led progressive search
(Handheld search)
Idea 3 – Question based search
(Handheld search)
Idea 4 – No search bar, entry via various routes (Handheld search)
Idea 5 – ‘Discover’ tab on navigation instead of search (Handheld search)
Idea 6 – Search only
(Mission-based)
(Mission-based)
Idea 7 – Search via social media and linking of accounts (Mission-based)
Idea 8 – Store scan search
(Mission-based)
(Mission-based)
Phase 2 – Refining concepts – From ideas to viable solutions
As we explored key user missions and refined designs to improve usability, discoverability and viability. Some ideas evolved, while others were parked for future consideration.
Evolution of concept 1 - Quick routes into search:
Initial wireframe – Concept utilises imagery to inspire and aims to provide a mix of searches based on trending items
Exploring search themes – A number of approaches were explored including: popularity, category focussed and newness
‘Inspire me’ tab – Category focussed search that crosses into other relevant categories
'Recent searches' entry points – Provide personalised and relevant content to enable but also inspire customers
Final concepts delivered – Search that guides and inspires
Search-enhancing concepts for both home and fashion categories were designed, prototyped and tested with users. Participant feedback was mapped against each concept, these supported our recommendations which were prioritised using the opportunity matrix and categorised into now, next, later and don’t do.
Detailed rationale, insights and recommendations were presented to stakeholders during our final playback of the project.
Concept 1 - Inspire me with quick routes into search (Before a search)
Insight: Some users click on the search bar without a specific product in mind. Personalisation (newness, popularity and offers) is appealing and can guide them towards discovery
Solution: An enriched typeahead displaying personalised, trending and content before users begin typing in the search bar
Key findings from user testing: Users found personalisation valuable but expected all content to be relevant and tailored
Opportunity – Richer typeahead: A/B test with different levels of personalisation to increase engagement and initiate a journey of discovery
Concept 2 - Search as a partner (Before a search)
Insight: Users often browse for inspiration but need structured guidance. Personable service is expected from John Lewis customers
Solution: A visually led, questionnaire-style interface, similar to the swipe feature in Tinder that refines search intent and tailors recommendations based on user missions and preferences
Key findings from user testing: Users appreciated the visual and interactive approach but expected all content to be tailored, personalised and relevant to their needs
Opportunity – Incorporate mission-based questionnaire and visual style suggestions: Balance guidance with efficiency, participants found being told which specific items go together valuable
Concept 3 - Search within a search with cross-shop opportunities (During a search)
Insight: Users are drawn to newness, popularity and offers, often without a specific product in mind. This behaviour leads users to click on the search bar without having a specific product in mind
Solution: Allows users to discover entire outfits or room setups directly from the search results, providing the opportunity to explore related items seamlessly
Key findings from user testing: Users appreciated seeing products in real-life contexts but needed clearer filtering, particularly for videos. The single-column layout felt overwhelming when navigating extensive product lists
Opportunity – Interactive shop the look, hero products on product listing pages and social media interactions (heart): Experiment with structured layouts that present all related products at once with styled or lifestyle imagery
Concept 4 - Rails of discovery (During a search)
Insight: Dead-end search experiences discourage inspiration and exploration and will stop "browsers' in their tracks. Users can feel overwhelmed by too many search results
Solution: Introduced filtering rails and alternative search prompts mid-search, expanding the user’s scope and lead them to discover more products, keeping users engaged longer
Key findings from user testing: Users appreciated the visual and interactive approach, engaging more when suggestions appeared within a few scrolls. However, they expected all content to be personalised and relevant to their original search term and specific needs.
Opportunity – Test Rails of discovery: A/B test as MVP across web and App
(A/B test on the app showed a +3.6% revenue per session increase)
Concept 5 - Integrating visually engaging content (During a search)
Insight: Social media and influencer content inspire younger audiences
Solution: Integrated influencer and user-generated content within search results leverages relatable and aspirational content that provides the personal expertise that customers expect from John Lewis.
User testing: Users liked the inspiration but struggled to gauge the relevance. Found the execution too busy and the use of 'Partner' caused confusion due to lack of familiarity
Opportunity – UGC & Stylist picks inspiration: Refine layout and blend content more seamlessly into journeys without cluttering
Concept 6 - Never-ending PDP (After search)
Insight: We don't help users continue their search journey after finding one product
Solution: Displayed dynamic product recommendations on product pages to keep users engaged post-search
User testing: Users expected product recommendations to be more contextually relevant to their original search
Outcome - Never-ending PDP: Enhance recommendation algorithms for better contextual fit to reduce dead-ends. Take into account the previous step of the journey
Summary of recommendations
Prioritised recommendations using the opportunity matrix, categorising them into Now, Next, Later, and Don’t Do to guide the next phases of implementation
Identified 4 MVP opportunities to test at scale: Never-ending PDP, Rails of discovery, Shop the look and Richer typeahead
Top 4 MVP opportunities
Richer typeahead
Build a ‘trending searches’ component within typeahead
Never-ending PDPs
Test existing ‘shop similar items’ component at the bottom of PDPs
Rails of discovery
Test ‘new in’ or ‘on offer’ rails below first fold of PLP
Shop the look
Test the shop the look feature across fashion products in browse
Green = MVP opportunities
Orange = Opportunities that showed promise but require more design thinking
Grey = Ideas that did not resonate with users in their current format and need reconsideration or redesigning
Business Impact & Results
A/B tests for the 'Rails of discovery' concept were tested across both app and web on Product Listing pages
2
A/B tests completed
3.6%
Increase in revenue per session
1.8%
Increase in conversion
£17.5M
Full year benefit
App
Saw significant results and re-ran the test to validate findings
Revenue per session increased by 3.6%
Conversion increased by 1.8%
Projected full-year benefit of £17.5 million
Web
Showed mixed results, prompting further refinement
Mobile:
Average order value: -2.7%
Average page views: - 1%
Desktop:
Average page views: - 1.2%
Next Steps
Draft ideas for future iterations flagged for further A/B testing to refine concepts
Integrate insights into product roadmaps
Explore ways to seamlessly blend inspirational content into search
Collaborate with designers to evolve concepts based on user feedback
My Reflections
This project validated that rabbit-hole searches can inspire users and drive engagement without disrupting findability. Our insights continue to influence product roadmaps.
Stakeholder management: Navigating a high-profile R&D project required strong stakeholder management. Weekly updates and presentations ensured stakeholders were engaged
User testing & Iteration: Rapid refinement of concepts and prototypes were key to defining viable user-driven solutions
Cross-functional collaboration: Close alignment with UX, PM and research enhanced learnings and execution
Balancing Inspiration & Usability: Ensured designs prioritised usability alongside engagement
Personal Growth: Learned new ways of working, strengthened UX/UI collaboration and applied learnings to future projects
Google Sites
Report abuse