ARKEN · Case study · Elisa Tron

TL;DR

ARKEN is a concept EDC I designed from scratch to answer a simple question: what changes when veterinary studies become the starting point instead of an edge case?

The problem

Veterinary studies vary by species, enrollment model, and protocol. Most EDCs handle that variation through custom workflows rather than product design.

My role

Sole Product Designer. Product strategy, UX, interaction design, design system, and front-end implementation from concept to working product.

The approach

Replace study-specific workflows with a configurable product model. Species, protocol, enrollment, and regulatory requirements become inputs instead of custom development.

What it unlocked

One platform that supports companion animals, livestock, individual subjects, groups, and multiple study designs without changing the underlying product.

Explore ARKEN →

The problem space

Most EDCs are designed for human clinical trials, then adapted for everything else.

Veterinary research doesn't fit that model. The same platform needs to support companion animals and livestock, individual animals and groups, species-specific assessments, and study protocols that differ in terminology, workflows, and regulatory requirements.

Instead of configuring a study, teams often end up adapting the product itself. Every new protocol introduces another exception, another custom workflow, or another feature built for a single study. What if variation was the starting point instead of the exception?

The product model

A platform that adapts.

Most EDCs treat every study as a custom product. New protocols introduce new forms, workflows, and exceptions until the platform becomes increasingly difficult to scale.

What I did was reduce every study to a small set of configuration decisions that generate the study structure.

Study type, species, enrollment model, blinding, and protocol settings become the inputs.

The goal was to create a system that could evolve with new studies without requiring new product design every time.

Study setup adapts as you choose: category, type, species, enrollment, and blinding each reshape the study.

Study configuration

How category and species shape the study

Each combination of category, species, and study type determines the enrollment model, whether a withdrawal period applies, and how blinding works. The platform derives these automatically from your choices.

Species	Typical study types	How animals are enrolled	Withdrawal period	Who stays blinded	Regulatory frame
Companion animal
Dog · CatRabbit · Other	TAS · Efficacy Bioequivalence Observational	Individual	None required Competition advisory only	Owner blinded CRC op.blind	VICH GL9 FDA NADA EMA
Horse	TAS · Efficacy Observational	Individual	None required Competition rules apply	Owner blinded PI op.blind	VICH GL9 FDA NADA
Livestock: individual tracking
CattleSheep · Goat	TAS · Efficacy Tissue Residue Bioequivalence	Individual Tagged within pen	Food safety · mandatory Slaughter / milk WDP Hard gate · no override	CRC op.blind Monitor unblinded	VICH GL9 FDA NADA USDA
Swine	TAS · Efficacy Tissue Residue	Individual Pen Tagged individual or pen aggregate	Food safety · mandatory Slaughter WDP Hard gate · no override	Farmer blinded CRC op.blind	VICH GL9 FDA NADA USDA
Livestock: group / pen tracking
PoultryBroiler · Layer · Turkey	TAS · Efficacy Tissue Residue	Pen Pen-level aggregate FCR · mortality · biomass	Food safety · mandatory Slaughter WDP Egg WDP independent timer	Farmer blinded CRC op.blind	VICH GL9 FDA NADA USDA
Livestock: population / herd tracking
CattleDairy herd · Cow-calf	Efficacy Observational	Herd Fluid population Animal-days at risk denominator	Food safety · mandatory Milking / slaughter Tracked at herd level	Site blinded Monitor unblinded	VICH GL9 USDA
FishAquaculture · Tank · Net-pen	TAS · Efficacy Tissue Residue	Herd Tank population Biomass denominator	Food safety · mandatory Harvest WDP Hard gate · no override	n/a	FDA CVM EMA · EFSA VICH GL9
Livestock: reproductive tracking
Swine · CattleDam and litter studies	TAS · Efficacy	Dam / Litter Dam + offspring cohort Offspring promoted at milestone	Food safety · mandatory Inherits dam WDP Colostrum exposure tracked	CRC op.blind PI op.blind	VICH GL9 FDA NADA

Enrollment model determines how the system counts subjects, calculates denominators, and assigns forms. · Withdrawal period is enforced as a hard gate at the database level for all food animals, not a warning. · Goat / Sheep frequently require ELDU (off-label) drug use; the platform flags this and requires a vet-set extended withdrawal with e-signature.

In practice

A walk through the platform.

Explore ARKEN →

An interactive tour of ARKEN: the study configuration, data capture, edit checks, the query workflow, and the audit trail across roles.

Impact

Measured by what the design made possible.

ARKEN isn't in production, so I can't measure adoption or business outcomes. Instead, I evaluate it by the product capabilities the architecture makes possible.

One product replaces study-specific workflows. Species, enrollment models, and protocol differences are handled through configuration rather than custom workflows.
Setup starts with a few configuration choices. The platform generates only the structure needed for that study, reducing complexity from the first screen.
Compliance is built into data entry. Out-of-range values become audit-ready queries at the point of entry, and every data change is fully traceable from the moment it's created.
A single study experience adapts to every role. Each role sees only the tools and information relevant to its work while sharing the same underlying data.

What I'll carry forward

The value of end-to-end ownership.

Owning the entire lifecycle meant every decision had to work across disciplines: intuitive for users, practical for engineering, traceable for compliance, and maintainable as the platform evolved.

Working across those boundaries reshaped how I think about product decisions. It gave me a deeper appreciation for the tradeoffs between user experience, technical architecture, regulatory requirements, and long-term scalability, making me a stronger partner to engineers, product managers, and domain experts.