From ‘Simon’ to Self-Serve: Redesigning Enterprise Commerce at Lumen
Lumen’s sales team had a nickname for the loading spinner on their quoting tool: Simon. As in, ‘Simon says wait.’ It appeared so often, stayed so long, that reps stopped fighting it and just named it. A loading spinner with a personality is a sign something has gone badly wrong.
I was lead product designer for two Lumen initiatives end-to-end: the internal quoting redesign that killed Simon, then the North Star design effort for Lumen’s customer self-service Marketplace — a 0→1 product built to compete with AWS and Azure for enterprise network infrastructure.
Project Description
Lumen had a legacy problem that ran deep. Three acquisitions — tw telecom, Level 3, CenturyLink — had layered complexity onto every system, every workflow, every team interaction. The quoting tool was a product of that history: location-first instead of service-first, because that's how the old infrastructure worked, not how sales actually sold. Every rep had to learn the tool's logic instead of the tool learning theirs.
The Marketplace was the next order of magnitude harder: build a consumer-grade self-service buying experience on top of genuinely complex enterprise networking products, across two separate design systems (Salesforce Lightning internally, Lumen Chi externally), with 10+ teams across UX, engineering, product, legal, marketing, and CX — while Lumen's executive leadership was betting the company's digital-first transformation on the outcome.
I was lead product designer, end-to-end, on both.
The Decision: Power Through Simplicity
The hardest design battle on this project had nothing to do with the quoting tool or the Marketplace build. It was convincing telecom product managers that hiding features was actually better product design.
The Dedicated Internet Access product had a problem I kept running into across legacy telco: everything was exposed. Every attribute, every config option, every edge case sitting right there on the screen. The reason was always the same — 'The DIA product is powerful because of its features. We have to show everything.'
But roughly 80% of those features were corner cases. Most reps, most of the time, needed maybe five options. The rest was noise — slowing configuration, increasing errors, making the product look scarier than it needed to be. I pushed to default to the common case, tuck edge-case configs behind a disclosure, and stop leading with complexity. That fight ran through the entire Marketplace.
Prototypes won it. Arguing about this in a meeting goes nowhere. Showing it — here's the current flow, here's the simplified one, click through both — that's how you move a room. Every contested decision on this project got resolved the same way.
Technologies Used
The Marketplace had to bridge two separate design systems: Salesforce Lightning for the internal quoting tool, Lumen Chi for the customer-facing buy flows. Different component libraries, different constraints, different everything. The goal was to make the switch invisible to users — even when it was very visible to the people building it.
-
Design & Prototyping: Adobe XD & UXPin
-
Design Systems: Salesforce Lightning (Internal Design System) & Lumen Chi (Customer-Facing Design System)
-
Usability Testing & User Research: UserTesting.com
-
Development & Project Management: Jira, Confluence, Salesforce, SAFe Agile/Scrum
What Actually Happened
-
Simon is dead. 70% reduction in time-to-task after the quoting redesign. Three things: flipping from location-first to service-first, refactoring the backend calls that caused the spinner, and defaulting to the common configuration case. None of them were dramatic on their own. Together, they changed how reps felt about opening the tool in the morning.
-
Prototypes did the persuading. Journey maps and spreadsheets were landing flat. I converted the research into full-fidelity, click-through XD prototypes and brought them into PI planning. Suddenly everyone was reacting to something real instead of arguing about something abstract. Alignment that would have taken weeks happened in a session. After that, prototyping before meetings was just how we worked.
-
The Marketplace shipped in under six months from North Star to launch. A year of North Star design work, then a sub-six-month production push. The dual design system piece — making Salesforce Lightning and Lumen Chi feel like one experience when you moved between them — was the hardest part technically. It shipped without visible seams, which was the whole point.
-
Still live. Still mine. By 2025, Lumen publicly reported 1,000+ customers on its Network-as-a-Service platform. The Marketplace buy flows still match the original mockups almost pixel-for-pixel. In enterprise software, that's genuinely unusual — product vision gets compressed and negotiated until what ships barely resembles what was designed. That this one held says something about how clearly the North Star was set.
Detailed Case Study
Enterprise SaaS
platforms shipped
design rework reduction
Parallels design system
task-flow reduction
Lumen self-serve
0→production
Parallels DaaS
Get in touch
I'm always excited to take on new projects and collaborate with innovative minds. If youhave a project in mind or just want to chat about design, feel free to reach out!