Leap: AI research assistant
Making source-backed AI answers feel clear, verifiable, and scalable as we expand data coverage and outputs (charts, audio, exports).
ROLE
PROBLEM
Users needed quick answers but also needed to verify sources and reuse outputs in their work.
SCOPE
• Answer reading experience (layout, hierarchy, interaction patterns)
• Onboarding and prompt guidance patterns
• Visual patterns for trust: citations, source presentation, and progressive disclosure
• UI behaviours for rich outputs (charts/infographics, audio playback, downloads/exports)
Overview
Leap is a conversational AI assistant that helps users ask questions and receive answers grounded in Mintel sources. Behind the scenes it uses a conversational Retrieval Augmented Generation (RAG) pattern, questions are translated/rephrased, relevant context is retrieved, and an answer is generated with references, with conversation history stored so users can continue later.
This work demonstrates my ability to design clarity in complex, data‑driven products: creating a scannable answer experience, making evidence easy to verify, and evolving the UI to support richer outputs (charts/infographics, audio playback and exports) without turning the interface into a cluttered dashboard.
Design judgement mattered because the product sits on a constant tension: users want fast answers, but also need to trust what they’re reading and trace it back to evidence, especially as answers become longer and more feature‑rich.
Before/after: The earlier answer experience pilot feature vs the updated, more structured layout.
The problem
Users needed quick answers but also needed to verify sources and reuse outputs in their work (docs, decks, stakeholder comms).
The experience broke down because:
As Leap expanded, answers became longer and harder to scan, increasing cognitive load and making follow‑ups less focused.
Trust depended on transparent sourcing: citations had to feel credible and usable, not just present.
We needed to introduce richer content (charts/graphs/infographics) and new formats (audio) without overwhelming the reading experience.
The interface had to work across screen sizes, remain accessible (keyboard/screen readers), and scale via repeatable patterns not one‑off screens.
Another old layout: As Leap scaled, answers became longer and more detailed. Without clear structure, key insights and actions were hard to locate, forcing users to read linearly and increasing cognitive load.
My role & collaboration
I owned:
UI and interaction design for the answer experience and related workflows (reading, verifying, acting).
Information hierarchy and layout model for balancing primary answer content with supporting evidence and utilities.
Design specs and implementation guidance to help engineering ship scalable UI patterns.
Visual indicators for feature rollouts and beta experiences so users understand when something is new or in test.
I partnered with:
Product managers to align scope and priorities with user impact (onboarding, response quality, adoption).
Engineering to validate feasibility early and iterate based on implementation realities.
Design peers to keep patterns consistent, accessible, and aligned with broader Mintel UI conventions.
How we worked:
Iterative collaboration rather than a single handoff lightweight reviews. This included rapid prototyping, and continuous refinement as features landed.
Design decisions
As Leap grew in complexity, I focused on keeping the experience clear, predictable, and easy to scale.
Make long‑form answers scannable without losing depth
What we were solving:
Users want comprehensive, insight-rich answers, but LLM responses typically appear as long, unstructured blocks of text. Because the model generates content sequentially, it doesn’t provide reliable hierarchy or prioritisation, making it difficult to quickly extract key takeaways or navigate the response.
What I designed:
I introduced a structured answer framework that layers information using progressive disclosure:
Collapsible sections break responses into manageable chunks
Consistent hierarchy makes answers predictable to scan
Persistent side rail keeps actions accessible without disrupting reading
Follow-up prompts support continued exploration
Structure is handled at the UI level, while content remains model-generated, ensuring consistency as responses scale.
Why this was the right call:
This approach makes long-form answers easier to navigate while preserving depth. Users can quickly identify key insights, dive deeper where needed, and continue the conversation without losing context.
By separating structure from content, the system remains flexible to different query types while maintaining a consistent interaction pattern.
Section hierarchy and persistent action rail (scrolling)
Bring charts into answers without turning the UI into a dashboard
What we were solving:
Users need evidence and data context, but embedding visuals can easily overwhelm a conversational reading flow.
What I designed:
Inline charts / infographics that appear when referenced, with interaction where appropriate, maintaining the written answer as the primary narrative.
Implementation notes:
Designed patterns that respect constraints (e.g. early limitations with copy & paste not including charts) and support progressive enhancement.
Inline charts within an answer, showing how it supports (not replaces) the narrative
Make answers feel trustworthy by showing where the evidence comes from
What we were solving:
For an AI answer to feel credible, users need to understand where the information came from especially when Leap is pulling in data-backed content like charts, category data, or other supporting sources. If that evidence isn’t visible or easy to follow, trust drops quickly. And when a question is better answered in another Mintel tool that Leap doesn’t yet surface directly, users still need a clear next step rather than hitting a dead end.
What I designed:
I focused on patterns that make the answer feel grounded and explainable; clearly surfaced citations and source links to show where claims or data points come from. Where Leap couldn’t yet feature a tool’s underlying data or codebase directly, I used a top-of-answer product router to guide users to the most relevant Mintel tool instead.
The impact:
This approach helped position Leap less like a chatbot and more like a research assistant that shows its working. Users could either validate what they were seeing through citations and sources, or be routed to the right destination when deeper product-specific exploration was needed. That balance supported both trust and continuity in the research journey.
Router banner at top of answer and cited supporting content within the response
Make the first-time experience self‑serve: prompt guidance, language, and accessibility
What I designed:
It became clear that many users weren’t especially AI or LLM-savvy yet. For first-time visitors, the challenge often wasn’t the interface itself, it was knowing how to begin, what kind of question to ask, or what a good prompt looked like. At the same time, returning users didn’t need hand-holding, they needed a faster way back into their workflow.
What I designed:
I introduced a homepage experience that treated first-time and returning users differently. New users were given lightweight prompt guidance to help them take the first step with confidence, while returning users were shown more relevant, usage-informed suggestions that helped them get back into the product faster.
The impact:
This helped turn prompt guidance into a genuine onboarding tool rather than just decorative UI. For less AI-confident users, it reduced the blank-page feeling and gave them a clearer entry point into the product. For returning users, it made the homepage feel more useful and responsive to how they already worked.
Homepage prompt guidance, first time visiting vs. returning user experience
Add audio playback to support accessibility and flexible working styles
What we were solving:
Some users may prefer, or need, to listen to longer answers rather than read them whether for accessibility reasons, multitasking, or reducing the effort of working through dense content.
What I designed:
An audio playback option (speaker icon) placed alongside existing actions in a consistent side panel, so it’s discoverable without adding noise to the answer body.
Implementation notes:
We iterated on supporting UI details (e.g., keeping feedback capture usable when actions move to the side, and aligning icon styling across the cluster).
Audio playback control in the side action panel
The outcome
The evolving Leap UI enabled richer, more answerable experiences by integrating additional sources while keeping the reading experience clear and trustworthy.
Evidence of impact:
Across the live product over 6 months, the same design direction showed up at scale:
~90% of users returned within 7 days of running a search, which meant strong repeat-use signal.
Feature adoption grew from ~21% to ~33% of all sessions, which meant users chose the new AI flow over alternative Mintel platforms.
In-product feedback ran ~2.4:1 positive, which meant users trusted the answers.
Reusable patterns:
The team gained a scalable answer layout model (reading-first + progressive disclosure) plus repeatable patterns for visuals, routing, audio and exports that continue to support iteration.
Reflection
This project reminded me that complexity doesn’t always need to be simplified away, it needs to be structured well. For AI-assisted products, trust is built through clear hierarchy, predictable interactions, and thoughtful disclosure of depth, so users can move quickly without losing confidence in what sits beneath the surface.