Streamlining Token Purchases

with an Integrated Buy Widget

Designed an in-app widget enabling users to buy presale tokens directly from their crypto wallet. AI-assisted tools helped us uncover usage patterns and accelerate research, allowing the team to focus on a smooth, trustworthy, and conversion-optimised presale experience from discovery to purchase and claim.

My Role: Design Lead / UX Designer
Duration: 10 months Team: 5 designers
Other Teams: Product, DEVs, Business, Data, CRM

01. Impact


  • Introduced a unified token purchase flow fully integrated
    within the wallet experience.

  • Improved user conversion and reduced friction across presale transactions.

  • Established a data-driven framework for continuous optimisation
    and post-launch iteration.

02. Overview


Challenge

Purchasing presale tokens required users to leave the wallet and complete transactions via browser widgets, creating friction, lowering trust, and breaking engagement. The goal was to design a seamless in-app experience that feels reliable, intuitive, and scalable for future tokens and chains.

Objectives

Create an end-to-end in-wallet token purchase experience.

Reduce friction from discovery to claim.

Reuse familiar patterns to ensure UX consistency.

Build a scalable foundation for multi-token and multi-chain expansion.

03. Research & Competitive Analyses


Benchmark

We evaluated leading browser-based buy widgets to understand functionality, accessibility, and friction points. AI-assisted clustering and summarization helped us:

Quickly identify recurring UX patterns.

Highlight where users struggled with connectivity or confirmations.

Spot opportunities for a connected experience, instead of an isolated purchase.

The findings revealed a key opportunity: most widgets treated purchase as an isolated action, while users needed a connected, end-to-end experience within their wallet.

04. Lifecycle Mapping


We mapped the entire presale lifecycle — from token discovery to token claim — to identify how this new feature could support users at every stage. This holistic approach helped define the value proposition:

This feature shouldn’t just enable purchase — it should guide, inform, and retain users.

Excerpt from the Presale Lifecycle — before, during, and after purchase

Key insights:

  • Many users arrive during high-urgency purchase windows.

  • Post-purchase steps (like claiming tokens) often lacked visibility.

  • There was potential to encourage engagement with new token listings.

05. Task Flow & Alignment


We created detailed task flows to align stakeholders on the shortest happy path, edge cases, validation steps, and feedback messages. We also considered emotional context, supporting users in high-pressure moments.

Task flow for happy path and other use cases

06. Wireframing & Iteration


We began with low-fidelity wireframes to outline layout logic and key interactions.
These were validated internally with cross-functional teams to ensure technical feasibility and content alignment.

Through multiple iterations, we refined the flow to balance:

Simplicity for first-time users

Familiarity with existing feature behaviours.

Efficiency in the fewest possible steps

Initial wireframes to align on UX intention

Flow mockup iterations

07. Edge Cases Design


Edge cases — such as insufficient balances, failed transactions, or unsupported tokens — were designed to provide guidance, not dead ends.

We focused on:

  • Clear recovery options and feedback messages

  • Reusing design patterns from other wallet flows

  • Maintaining trust and retention during error handling and retries

08. Delivery & Handoff


Once approved, the designs went through an engineer review to validate feasibility and interaction logic before handoff. We then finalised components and delivered production-ready files with complete implementation documentation, including:

Complete handoff file with all states and flows

Screen-level annotations for default and interaction

Detailed specs for integrations with backend

09. Design QA


Before release, we ran a full Design QA (Quality Assessment) to ensure implementation accuracy. Each issue was categorised by impact:

High

Blocks user flow

Medium

UX deviation, recoverable

Low

Visual inconsistency

This systematic QA reinforced the product’s design quality and ensured parity between the final build and the intended user experience.

10. Post-Launch Behavioural Insights


After launch, we collected feedback, analytics, and session replays across hundreds of sessions. AI tools helped reveal hidden patterns — where users struggled or dropped off — and guided targeted UX refinements.

We reported these insights weekly to product and business teams, highlighting behavioral trends and friction points that weren’t visible in standard quantitative data. These insights informed design tweaks, improving conversion, clarity, and overall flow efficiency.

Session replay analysis, observation notes and a presentation of insights to the team.

11. Continuous Improvement


Several iterations followed post-launch, focused on:

Reducing friction during conversion and error recovery

Supporting new chains and token integrations

Improving clarity in status and confirmation feedback

These improvements led to a measurable increase in user conversion and transaction completion, reinforcing the value of continuous data-led iteration.

12. Conclusions


This project demonstrates how strategic UX can combine technical complexity, business goals, and user trust in a single, coherent experience. By embedding user insights and AI-assisted analysis at every stage, design became a continuous driver of adoption, conversion, and long-term engagement.

Design doesn’t end at launch — it evolves with every user interaction.

Previous
Previous

Creating a human-centered app for home cooks.

Next
Next

Connecting Portugal and Brazil through wine and design