Pharma teams often need to create brochures and presentations that explain how a new drug works — usually for healthcare providers or internal reviewers. These materials must include claims backed by clinical trial data, like how effective the drug is, what side effects it has, or what outcomes were observed.
The problem? That info is buried in hundreds of pages of dense, technical documents. Finding it is time-consuming, tedious, and easy to mess up — especially when accuracy and compliance really matter.
I helped design an AI-powered tool for Eversana, a healthcare company that helps pharmaceutical brands bring new treatments to market — who needed a faster, more reliable way to build these materials without risking errors.
The tool uses natural language processing to scan clinical trial documents and pull out key claims based on medical language patterns. Each claim is tagged with its exact source — like the page and line number — and saved to a searchable database.
It also reviews external-facing materials used to promote a drug — like brochures, sales decks, or leave-behinds for healthcare providers — and automatically highlights the claims being made. Then it links each one back to the verified data in the original trial docs, so teams can be confident everything they’re saying is accurate and backed by evidence.
The result? Less time wasted, lower risk of compliance issues, and materials teams could actually trust.
Pretty cool, right? (hopefully the answer is yes)
I was the only designer on the TensorIoT side, responsible for turning a set of requirements into a real, usable product. While I was given a general visual direction and a list of business needs, I handled everything from workflows and layouts to interactions and detailed UI.
That meant designing the full experience — including navigation, tables, and how users would move through the tool to upload documents, review claims, and access data.
I worked closely with Eversana’s UX designer and business analyst to make sure we were aligned on both business and user goals. Since their internal UX team was our direct client, every design first had to pass their review before going up the chain to the VPs of Marketing and Technology.
Early on, I flagged this two-step approval process as a possible risk to our timeline. I raised it with our project managers and suggested building in buffer time — which paid off later when delays came up. That foresight helped keep the project moving and built trust across the team.
A diagram highlighting my impact in the process. Tap for a larger view!
(if you'd like to call it that, I enjoyed working on this project immensely, and genuinely found the obstacles great opportunities to learn and be better. I will admit, working with great people made it a very supportive process.)
Company list
Business unit table (organized by company)
Brands (associated with Business unit and Company)
HopeWell brand dashboard
Eversana planned to offer this tool to multiple pharmaceutical clients, so the platform needed to serve many companies at once — while keeping each company’s data private and secure. This is known as multi-tenancy: one tool, shared infrastructure, many companies.
On top of that, Eversana’s internal teams needed to switch between companies, manage their data, and assign access to different user roles — all from a single dashboard. That added complexity across the product — from navigation to permissions to what each user could see or do.
We were designing for five user roles:
1. Eversana SuperAdmin: edit access to all customers and brands.
2. Eversana Admin: edit access to certain customers and brands.
3. Eversana’s External Customer Admin: edit access to their own brands.
4. Eversana User: read-only access to certain customers and brands.
5. Eversana’s External Customer User: read-only access to their own brands.
(“External” here means users who work at the pharmaceutical companies using the tool.)
To support this structure, the product needed to represent a hierarchy:
- Companies = Eversana’s pharmaceutical clients
- Business units = teams within those companies focused on areas like oncology or cardiology
- Brands = the actual drugs
- Indications = the conditions those drugs treat
Eversana’s original idea for multi-tenancy
Eversana's original idea was to show all of that on one screen so users could manage the full hierarchy in one place. I flagged early that this was probably out of scope. To confirm, I brainstormed alternatives and worked with developers — and we agreed it wasn’t feasible for this phase.
After we validated the limits, I led the redesign of a simpler but scalable version — one that broke the experience into steps, matched each user’s access level, and could grow in future phases. You can take look at the final design in the carousel above!
The Eversana team appreciated my ability to simplify this complex system without compromising usability. By identifying scope risks early, collaborating closely with developers, and offering a scalable alternative, I helped them avoid costly delays and maintain stakeholder trust. This proactive, transparent communication kept everyone aligned — and gave them confidence moving into the next phase.
Due to budget and timeline constraints, we weren’t able to run formal usability testing during this phase. So instead, I focused on designing with built-in clarity and flexibility, making sure every screen made sense even without guidance.
We used regular design reviews with Eversana’s internal UX team as informal checkpoints. Since they were closely aligned with their customers’ needs, they helped flag potential usability issues and made sure the experience worked across different user roles.
I also made sure the designs were demo-ready, so their internal team could present the tool to clients and start gathering feedback once it launched.
Real user insights are expected in the next phase, once the tool is live and in regular use. The team plans to run usability testing and make improvements based on how it performs in the real world.
If you’re curious how I typically approach user testing when time and budget allow, this case study shows what that process looks like.
Even though we ran a few weeks past the original deadline (mostly due to unexpected development challenges), I made sure Eversana was kept in the loop the whole time. I shared updates early and often — which helped build trust and keep momentum steady. That kind of transparency is something I always bring to my projects.
During the final project retrospective, Eversana shared that they were impressed with how quickly I picked up the complexity of the pharmaceutical space — especially since I had no previous experience in it. They appreciated how I asked the right questions, listened closely, and translated their business needs into clear, scalable designs.
It was also great to hear that they were happy with the final product and excited to keep moving forward. Their positive feedback led to early conversations about a potential next phase — which is always a nice sign.
