US Wealth Management Firm · May 2023 - Jul 2026

Rebuilding the Checklist at the core of a wealth management platform

As lead product designer for over three years, I reworked the Checklist advisors relied on throughout every client profile - not a one-time setup step, but the hub they returned to again and again - then built out the design system, recommendation tools, and analytics dashboards on top of it.

This project is under NDA - the company name and product branding have been withheld, and all screens shown have had logos removed.

Team

Embedded design partner to the COO, CTO, and product managers

Scope

Checklist, design system, portfolio recommendations, analytics dashboards, reporting

Constraint

Every workflow had to stay accurate for regulated, real-money tax and portfolio calculations

Duration

May 2023 - Jul 2026, continuous iteration based on customer feedback

The problem

When I joined, the platform's Checklist - the hub advisors returned to again and again as they worked a client profile, not a one-time setup step - was supposed to be the core of the app, but it wasn't working. It asked for too much at once, the process wasn't fluid, and advisors kept getting stuck on things like document processing and statement extraction every time they came back to it. There was also no design system to speak of - just a set of fonts and colors the client liked - so nothing about the interface stayed consistent as new features got added. Over the next three-plus years, working directly with large advisory firms serving high-net-worth clients, we rebuilt the Checklist experience from the ground up, then used that same foundation to build out the platform's recommendation tools, analytics dashboards, and reporting, tackling specific parts as customer feedback came in and new features got scoped.

Process

Rebuilt the Checklist around granular, contextual steps instead of one dense page

The Checklist wasn't something advisors finished once and moved past - it was the hub they kept returning to throughout a client profile, and every time they came back it asked for too much at once. I broke it apart into smaller, contextual steps that surface right where an advisor already is, so they always see one clear next action instead of a dense page and dead ends like a document that wasn't parsing correctly.

Built the platform's first design system, then made it legible to AI agents

There was no design system when I started - just a set of fonts and colors the client liked. I built a full Figma system covering typography, color, tokens, spacing, and a component and icon library, then wrote a Design.md file distilling it into something AI coding agents could reference directly, since AI-assisted development was already becoming part of how the team shipped.

Used AI to prototype faster across every new feature, not just the big ones

I used to build Figma prototypes the old way, wiring up interactive components and variants by hand. Once tools like Figma Make, Replit, and Lovable became part of my process, prototyping got roughly ten times faster - which meant more directions in front of stakeholders before committing to one, instead of betting early on a single approach.

Selected screens

The original inherited Checklist, broken into Investment Data, Analysis, and Proposal steps, with product branding blurred — The Checklist I inherited - already broken into steps, but still asking for too much at once every time advisors came back to it, with dead ends like stalled statement extraction.

The reworked approach - instead of one dense Checklist page, the relevant next step surfaces as a contextual prompt right on the dashboard, wherever the advisor already is.

The Profile Dashboard once data is in - fitting everything advisors needed into one structured view, without the overlapping, freely-resizable widgets stakeholders originally wanted.

Portfolio Recommendation Creator mapping current accounts into proposed recommendation accounts via drag and drop, with product branding blurred — The Portfolio Recommendation Creator - drag accounts, securities, or specific tax lots between current and recommended, splitting or transferring amounts with calculations updating live.

Tax transition model showing lot-level trade values and computed tax costs against advisor-set constraints, with product branding blurred — The tax transition model inside the Recommendation Creator - real constraints in, per-trade tax cost out, at the tax-lot level.

Key decisions

Convinced leadership the Checklist was the product, not a gate in front of it

For a while, the Checklist was treated as a formality advisors passed through once on their way to the 'real' product elsewhere. I argued the opposite - advisors were going to live in this hub throughout a client's entire lifecycle, so a broken experience there meant a broken experience everywhere, every time they came back. Getting the COO and CTO to rework the core concept around that idea was the single biggest unlock of the whole engagement; once that shift happened, everything downstream went easier.

Pushed back on resizable, overlapping dashboard widgets

Stakeholders wanted the Profile Dashboard to support freely resizable, overlapping widgets so advisors could customize their view. I pushed back - that flexibility would have made the dashboard look inconsistent and cluttered for most advisors, for the benefit of a small number of power users. We kept a structured, opinionated layout instead, prioritizing a dashboard that looked right by default over one that was maximally configurable.

Outcome

Contextual, not dense

the Checklist redesigned around small, always-forward steps advisors could pick up anytime, instead of one dense page

Design system, from zero

full Figma system - typography, tokens, components, icons - plus a Design.md spec for AI agents

~10x faster prototyping

AI-assisted tools cut prototyping time enough to explore more directions before committing

Described qualitatively - exact adoption and efficiency figures weren't tracked in a form precise enough to cite here.

What I’d do differently

“Three years with the same stakeholders means you win some arguments, lose others, and a lot end up as compromises - that's normal. What actually mattered was picking the right fight. The checklist redesign was the one decision I pushed hardest for, because I could see it was the thing blocking everything else, even though it wasn't the most visually interesting problem on the table. Most of what I'd do differently next time is really just a version of: find that leverage point earlier, and spend less energy on the arguments that don't move the whole product.”

Next case study

Designing the analytics platform behind global nuclear safeguards verification