BACK
19:02:17 UTC
Lexicon is a Figma plugin that helps biotech designers validate language, access domain knowledge, and stay aligned with scientific accuracy.
Case study
Designers in biotech often face a steep knowledge gap. Without a background in science, even experienced designers struggle to interpret and apply technical content in their work. Lexicon is a Figma plugin built to help them speak the same language—literally.
INTERVIEWEES FROM
In biotech organizations, designers often work in unfamiliar scientific territory.
Without a clear grasp of technical concepts, they risk misrepresenting content, miscommunicating with engineers, or slowing product development altogether. But scientists don’t always have time to translate complex concepts for them, and traditional documentation — scattered across PDFs, Notion, and Slack — isn’t built for real-time reference.
This wasn’t a documentation problem. It was a communication breakdown embedded in tooling, structure, and time.
Design, engineering, and science each spoke their own language. The real challenge was enabling designers to operate fluently across disciplines without requiring them to become experts.
And the solution couldn’t be another system to learn. It had to work inside the tools they already used.
This wasn’t a documentation problem. It was a communication breakdown embedded in tooling, structure, and time.
Design, engineering, and science each spoke their own language. The real challenge was enabling designers to operate fluently across disciplines without requiring them to become experts.
And the solution couldn’t be another system to learn. It had to work inside the tools they already used.
I talked to 11 Designers & Design Leaders in Biotech, and three core issues emerged
Domain knowledge was hard to access
Designers relied on scattered chats, legacy documents, or overburdened coworkers to answer basic terminology questions.
Scientific language lacked structure
There was no consistent way to reference or validate technical terms in design. Unlike typography or color, language wasn’t treated as a design system.
Onboarding was inconsistent and overwhelming
New designers spent weeks figuring out what words meant and how to use them. There was no way to surface relevant terms in context.
To address these issues, I focused the solution on three principles
Cross-disciplinary
Designers and scientists don’t need to become each other, but they do need a shared language. This was about lowering the barrier to understanding without oversimplifying the science.
Consistent
We already treat type and color as systemized. Why not language? Building consistency into terminology helped designers work faster and communicate more clearly.
Contextual
No one should have to stop their flow to Google a term. The tool needed to answer questions right where they came up—quietly, instantly, inside Figma.
But I didn't jump straight to building
Before moving into design, I looked at how language issues were actually showing up in the workflow.
Often, they appeared in messages: short, informal questions that signaled real uncertainty:
These weren’t edge cases. They were repeated signals that designers didn’t have access to the terminology support they needed, when they needed it. And while these questions seemed small, they slowed down reviews, increased reliance on engineers, and made onboarding harder than it needed to be.
I traced when and where these moments were happening, handoffs, design reviews, copy updates, and used these touch points to define key points of intervention.
This shift from documentation to embedded support shaped how the design idea came together.
These conversations surfaced three key insights:
Designers weren’t confused by the design, they were uncertain about the language.
The friction wasn’t visual. It was semantic. Even experienced designers hesitated when they weren’t confident in the terms they were using.Terminology questions didn’t just happen at onboarding, they happen mid-flow as well.
By the time someone asked for clarification, they were already deep into a task. The system needed to support just-in-time understanding, not just up-front training.Asking questions slowed down teams, not because they were complex, but because they were frequent.
These weren’t rare edge cases. They happened daily. The cost wasn’t in the difficulty but in the accumulation.
The scope is clear: we need to remove ambiguity at the exact points that's blocking progress.
The Solution: A Figma Plugin That Thinks Like a System
Automated Text Validation
As a Figma private org plugin, Lexicon allows organizations to establish their own design language library – just like color styles and text styles.
Referencing a set of language styles, the plugin detects if language in design is correct and appropriately placed, allowing designers to ship more accurate designs and accelerate the product development process.
Customizable Language Library
The language library itself is composed of different parameters, which are used as reference for the text validation feature. These parameters are fully customizable, such as ideal placement, purpose for using the language, and some relevant assets like appropriate icons.
Unified Knowledge Base
Just like a company wiki, the plugin allows designers to search through the company's knowledge base when encountering questions.
This feature allows cross-platform integration, enabling designers to search through all platforms at once without having to navigate across 10 different Apps.
Designing for trust, not just terminology
Each feature in Lexicon was designed to resolve a specific point of friction in the design process. Not by adding more rules, but by making the right patterns easier to follow.
Instead of overhauling workflows, Lexicon embedded guardrails into places designers already worked. Validation made language feel safer. Structured libraries reduced ambiguity. Search turned interruptions into answers.
Together, these changes helped designers ship more confidently, align faster with technical teams, and stop second-guessing the words they used.
Before
after
Terminology
Unstructured, undefined, often incorrect
Defined, validated, scoped to usage
Workflow
Dependent on coworkers, context switching
Self-sufficient, embedded in design process
Onboarding
Slow, inconsistent
Guided by in-context documentation
Takeaways
Embedding validation directly into design tools builds trust without disrupting flow.
Standardizing scientific language empowers cross-functional teams to move faster
What feels like inconsistency is often a signal the system needs to adapt
Good systems don’t enforce knowledge—they surface it at the right moment
SEE ALSO