Case study

Reimagining Digital Investing with AI-Powered Guidance

Self-directed investing platforms give users powerful tools, but little guidance when it matters most. This project reimagined the investing experience by embedding AI-powered guidance directly into key decision moments, helping users move from uncertainty to confident action.

My role focused on owning the end-to-end design of key features, translating the vision into user flows, wireframes, and high-fidelity prototypes. I drove decisions around interaction patterns, information hierarchy, and how guidance is integrated into critical user moments.

My Role

  • Wireframing
  • User flows
  • High-Fidelity UI

Team

Lead designer, UX designer and me (UX Designer)

01 — The Challenge

How might we help users make better financial decisions without taking control away from them?

Self-directed investing platforms give users control—but also expose them to risk, uncertainty, and cognitive overload. Through early discovery and domain research, we identified a set of recurring friction points:

Portfolio changes feel opaque and difficult to interpret
Trade decisions lack clarity (risk, diversification, tax implications)
Educational content exists but is disconnected from real decisions
Emerging investors expect intuitive, guided experiences — not just tools
02 — Framing the Opportunity

Users don't want automation. They want confidence.

We reframed the platform from a toolset to a guidance system. Instead of replacing decision-making, we focused on:

Supporting users with contextual insights
Increasing awareness of consequences
Preserving autonomy at every step
03 — Approach

We worked iteratively, using lightweight prototypes and storytelling artifacts to align stakeholders and refine direction.

Key Activities Included:

Collaborative workshops to align on vision and scope
Journey mapping across onboarding, portfolio, and trading flows
Rapid white-boarding and concept iteration
AI prototyping to explore conversational and contextual interactions
Development of experience principles to guide decisions

We focused on translating abstract AI concepts into clear, actionable user interactions that could be understood in high-stakes moments.

04 — Users

We started from two user profiles defined by the client. From there, I identified the shared needs and key tensions that guided the design decisions.

👴🏼

Less digitally confident investors

  • Lower confidence in digital tools
  • Higher risk aversion
  • Prefer clarity and reassurance
👦🏻

Digital natives generation

  • Comfortable with autonomy
  • Faster decision-making
  • More prone to overconfidence and impulsive trades
Shared need: Understanding the consequences of their actions before committing.
05 — THE SOLUTION

From transaction-first investing To Guided, confidence-building decision-making

We designed a hybrid AI guidance system embedded across the investing journey.

Instead of a single feature, the solution works as a layer of intelligence that adapts to user context.

Embedded AI assistant interface showing contextual guidance across the investing platform
01

Embedded AI Assistant

Goal: Make guidance always available, but never intrusive.

Explains decisions in plain language
Answers contextual questions
Builds trust through transparency

Instead of isolating AI in a chatbot, we embedded it throughout the experience

Embedded AI assistant interface showing contextual prompts and plain-language explanations within the investing platform
02

Pre-Trade Intelligence

Goal: Help users act with confidence

Executing a trade is one of the highest-stakes moments in the experience. We designed a pre-trade layer that surfaces:

Trade impact summaries (risk, diversification)
Scenario-based insights (what changes if I proceed?)
Contextual explanations tailored to the user's portfolio
"How will this affect my portfolio balance?"
"What should I consider before making this trade?"
Pre-trade intelligence screen displaying trade impact summaries, risk analysis, and scenario-based insights before executing a trade
03

Learn by Doing

Goal: Create a safe environment for experimentation.

A safe environment where users can simulate strategies without real risk.

Why it matters:

Reduces fear of making mistakes
Encourages exploration and learning
Sandbox portfolio simulation interface where users can practice investment strategies without real financial risk
04

Scenario Exploration

What If?

Goal: Help users explore "What If" outcomes before committing.

Design Decision:

A tool that lets users explore outcomes before committing.
Impact: Users shift from reactive to proactive thinking
What-If scenario exploration tool showing projected portfolio outcomes based on different investment decisions
06 — Roadmap & Mpd definition

To bring the vision to life, we defined a phased approach focused on delivering value at key moments in the investing journey — rather than launching a fully comprehensive AI system all at once.

PHASE — 1

Guided Foundations

Focus: Build confidence in core user journeys.

Key Features:

Portfolio health overview (asset allocation, diversification, risk, ratings).
Plain-language explanations of portfolio performance.
Contextual prompts introducing AI-assisted guidance.

Outcome: Users gain a clear understanding of their portfolio and what to do next—without feeling overwhelmed.

PHASE — 2

Decision Support

Focus: Support high-stakes moments.

Key Features:

Pre-trade insights (impact on risk, diversification, alignment to goals).
Scenario exploration ("what happens if I make this change?").
AI-assisted explanations tied to user intent and portfolio context.
Guided flows for making portfolio adjustments.

Outcome: Users move from uncertainty to informed action with greater confidence.

PHASE — 3

Proactive & Personalized Guidance

Focus: Anticipate user needs over time.

Key Features:

Intelligent alerts tied to portfolio health and market events.
Personalized recommendations based on goals and behavior.
Ongoing learning and education embedded in the experience.
A more adaptive AI assistant that evolves with the user.

Outcome: The platform shifts from reactive toolset to proactive financial partner.

Rather than treating AI as a standalone feature, this roadmap positions it as a progressively deepening layer of guidance — starting with explanation, expanding into decision support, and ultimately enabling personalized, proactive advice.

07 — Outcomes

Turning AI Vision into Actionable Product Direction

  • Translated a complex AI vision into tangible product experiences
  • Enabled stakeholders to align on a clear direction for AI-assisted investing
  • Defined a realistic MVP and phased roadmap
  • Created high-fidelity prototypes to validate and communicate high-stakes interactions
  • Established reusable patterns for AI-assisted decision-making
Final high-fidelity mockup of the AI-guided digital investing dashboard showing portfolio overview and guidance features
07 — Key Takeaways
01

Guidance is most effective when it respects user autonomy

Users don't need more tools — they need help using them

02

AI works best when embedded, not centralized

Context matters more than capability

03

Confidence is a design outcome

Helping users feel confident is just as important as functionality

09 — What I'd Improve Next

Always room to grow

east Personalize guidance based on behavior over time
east Validate with real users in live trading environments
east Further reduce cognitive load in high-stress moments
Closing Thought

Designing for financial decision-making isn’t just about usability—it’s about responsibility.

The goal isn’t to make decisions for users, but to help them make better ones.

Improving Efficiency in a Customer Support Operations Tool
arrow_forward