Let me tell you about how we transformed Soda, a SaaS platform in the PDF utilities world.
Their technology kept moving forward, but the interface stayed stuck. The team called it “legacy jail”. And over time, the gap between capability and UX compounded scattered, stacked inconsistencies, creating a gap between what the product could do and how it actually felt to use it.
And that's where I stepped in.
My role: Ux audit, product design, alignment coordination.
Context
The first thing I noticed, is what I call The Production Trap: a team working in isolated tasks, generating high-detail UI for ad-hoc updates, with no strategic shared point of reference. Designers were creating bottlenecks without knowing it, and devs… well, were confused.
Having learned that process is just as important as output, my suggestion was to conduct an audit, prioritize updates by impact, and tackle them globally.
Process
It all started with a usability audit, and competitor analysis I took on, to surface what wasn't working and where the real opportunities lived, which gave us a baseline to take decisions.
Once the improvements surfaced, we grabbed the low-hanging fruit items that wouldn't ripple into functionality to set a clean baseline. The goal became to expose the user to the UI, and then reevaluate the UX without the visual noise getting in the way.
Results
After building a set of guidelines backed by a small sample set of screens, a structure emerged that could support future updates and potentially become the foundation of a design system.
The improved process gave developers clarity on what needed to be adjusted and what wasn't necessary, reducing back and forth and allowing them ship faster, which freed the team to get creative in their roles.
As for the product, the improved consistency across patterns reduced visual noise. The results showed up as lower cognitive load, faster task completion, better net promoter scores, and a shift in how users talked about the product.
What the initiative covered
The project involved a mix of heuristics, process alignment alongside PMs and the dev team, while creating a set of prototypes that documented patterns so collaborators could visualize the improvements and create tasks.
Day to day, the work became:
Conducting a usability audit grounded in heuristics and industry best practices.
Benchmarking, studying what competitors do, to understand gaps and inform the strategy.
Getting stakeholders aligned through workshops and UX artifacts.
Finding the “low-hanging fruit”, fixes that wouldn't disturb the tech.
Partnering with PMs and POs to scope and create tasks.
Using AI to synthesize scattered company resources into documentation.
Creating prototype sets.
Built right but “felt” wrong
Users and internal feedback consistently pointed to a feeling of slowness or something being "off." In isolation it seemed ambiguous, and stakeholders couldn't pinpoint the cause.
Looking at it as a whole, a pattern emerged: the tools were functioning as expected and followed requirements, yet the combination of overlapping shadows, transparencies, embellishments, and color mismatch was accidentally “mimicking bugs”. On top of that, a lack of visual feedback on completed actions left users unsure whether the tool reacted to them.
All of this created a disconnect on how the product felt and how it actually performed, undermining trust.
Here you can see how the interaction feedback issue got fixed.
Previously the tool controls had only Default, and Hover states, leaving people uncertain whether their clicks registered. The disabled states looked similar to enabled ones, which was problematic too.
The redesign introduced five distinct visual cues that gave users immediate clarity on what action was taking place.
The central challenge.
Like most products built at speed without a system, the interface evolved feature by feature, with different designers shaping parts independently. No version held it together and contradictions piled up, turning what could’ve been updates into archaeological digs, which slowed projects down.
A usability audit and mapping the system gave the team reference points, something to orient themselves within the product. It turned Devs and PMs into advocates of strategy, while prototype sets showed the nature of the updates, which ensured quality.
Here another example: The old navigation forced users to dig through multiple layers and infer which tools belonged together and what you could actually do with them, making it tedious to grasp how the tools connected.
When we upgraded the style I introduced a tab system, that had a clearer differentiation on the "mode" the feature was in, so jumping between sections naturally showed you which tools belonged together.
Final thoughts
The usability audit, and benchmark helped identify where the experience was falling short. Research insights then pointed at the moments of friction. Together, these narrowed down the updates that would deliver the most with the least effort.
Visual noise during key processes can create the appearance issues while masking real problems, and when that layer sends the wrong signals, like in our case here, users take the mental burden, work harder to interpret what's happening.
By removing certain things and adjusting others, the UI started reflecting the actions taken. And we got less and less “bug” pointers without actually touching the tech. In other words, the new user feedback became more actionable.
I appreciate you reaching here, now… excited for what's next?