Overview
Factor Eleven runs an advertising platform, and its finance team was tracking the numbers that mattered most, like burn rate, MRR, CAC and runway, across a pile of spreadsheets and legacy tools. Everyone had a slightly different version of the truth.
As the sole product designer, I took the project from the first discovery conversations to a developer-ready dashboard: research, wireframes, iterations with stakeholders and a clean handoff.
The result was one self-serve finance dashboard with access levels, so everyone who needs the numbers can see them instantly, without waiting on finance.
The impact
How it was before
Spreadsheets everywhere
Key metrics lived across 9 spreadsheets and legacy tools, each updated by hand. Preparing a report took 2 days.
No single source of truth
Teams read the same numbers differently, so meetings started with "which version is right?"
Finance as a bottleneck
Getting one number could take 2 days and a few people to ask, so every decision waited on finance.
On top of that: no existing design foundation, complex financial data, very different needs across departments, and the limits of existing systems.
The goal
Give the whole company one honest view of its financial health: accurate and up to date, readable by finance people and non-finance people alike, and ready to scale as new data and teams came in.
Success meant fewer "can you pull this number for me?" messages, and decisions made from the same data.
How I approached it
- 1
Discover
Interviews, shadowing and an audit of every report, to understand how decisions were really made.
- 2
Explore
Wireframes of two layout systems, reviewed with finance, leadership and engineering in tight loops.
- 3
Deliver
High-fidelity UI, reusable components and an annotated, developer-ready handoff.
Listening first
I ran 1:1 interviews with finance, operations and leadership, shadowed the weekly reporting routine and audited every report and tool the team relied on. I also benchmarked ChartMogul, Baremetrics and the internal dashboards of Stripe and Intercom.
The same three problems kept coming up: too much information, no clear sense of trends, and no shared understanding between teams.
- CFO & finance team — for data accuracy and metric definitions
- CEO & leadership — to understand how decisions actually get made
- Engineering lead — to align on feasibility and data constraints
Two layouts, one answer
My first wireframes explored two directions: a modular card system with drag-and-drop flexibility, and a page-level hierarchy with grouped zones like Revenue, Spending and Efficiency.
Stakeholder reviews pointed to a mix of both. I kept the clarity of grouped zones and built them from reusable components, so the dashboard could grow without being redesigned. Feedback ran through Figma comments, async Loom walkthroughs and weekly design syncs.
The MVP journey
The solution
Everything that matters, at a glance
Quick facts and real-time revenue sit at the top, so anyone can see how the business is doing in seconds. Export, refresh and filters are always one click away.
Zoom in, zoom out
Leadership can scan the big picture while campaign managers drill into goals, billing types and agencies without switching tools.
Your data, on demand
Dynamic date ranges and filters by country, media product and campaign type, with export built in, so nobody has to ask finance for a custom report again.
Details that matter
Colour-coded KPIs
WCAG-compliant colours make risky numbers easy to spot without relying on colour alone.
Trends with context
Trend lines come with short explanations, so a dip reads as information, not panic.
Collapsible detail
Secondary data folds away, keeping the default view calm and focused.
Access levels
Each role sees the numbers it needs, so data can be open by default without exposing everything to everyone.
Edge cases covered
Empty states, loading and long values were all designed and annotated before handoff.
Final designs
I delivered a developer-ready Figma file: organised components, defined styles, reusable KPI cards and data visualisations, and every screen annotated with usage notes and edge cases.
What I took away
Dashboards fail when they try to show everything. The biggest wins here came from deciding what not to show by default, and making the rest one click away.
Building on reusable components from day one also paid off: role-based views and new data layers can be added later without starting again.



















