Search engine for chronic diseases
Project goal
The goal is to allow an overview of all current business partners of a pharmaceutical Swiss company working with a chronic disease, via the connectivity of the existing databases.
My contributions
As a UX Designer, I conducted user generative interviews, analysed and prioritized the results to create use cases. Created high level user journeys and 1 high fidelity clickable prototype in Figma. For testing I conducted usability tests, gathered feedback and iterated 3 times the design concepts and created 3 new future feature concepts.
The product
Disclaimer: Due to NDA constraints, I am unable to showcase all real screens, similar data or reveal more of the company involved in this project. All examples and details have been modified and anonymized to adhere to these confidentiality requirements.
Search result view
After searching for a chronic disease, users can see which companies work in this disease and relevant data.
This solved the need to find the most relevant data at first glance.
Search result view with slider open
A slider with a 2nd level of more detailed view was introduced because some users needed to learn more about a partner they were interesti Products, Partnership and who to contact about this company.
Detailed views from a company
A third level of information was introduced where users could download documents and learn about their current or past collaboration together.
What problem did we solve?
With machine learning, PDF's, Word documents and Google sites were scraped to put together the information shown by the search engine, also making these available for download.
This process lowered the user’s average time for desk research from 2 weeks to some minutes. This saved employees days and weeks of waiting times/work.
The process
01
The problem
In the exploration phase, we discovered our client was having issues with finding the right documentation to fulfill their tasks.
The documentation was made in Google sites, Excel files, PDFs and Google docs. There was not a single source of truth. Finding it was hard and tedious.
If a user wanted to do a search of a chronic disease or partners working on it, they would have to refer to Google Drive, call/email colleagues and directly ask other departments if they had the most available and recent documentation needed.
This process was highly manual and would take up to 2 weeks to finalize.
We identified pain points:
- "Sometimes I don't know who to ask about this information I need"
- "I can't find what I need"
- The waiting times for a search were unpredictable, as well as the completion of the process
- "I don't know how to find all these documents"
02
Use case and High level user journey
"As a User I would like to get an overview of all potential partners working in the area of a chronic disease." (Use case)
I visualized 6 high level current-state customer journeys on Miro that were discussed with the stakeholder for accuracy purposes. Most of the users had the same role, but interestingly their tasks and processes were different to achieve the same goal.
03
Prioritisation of content and parameters
We had a workshop with the result of prioritizing 8 parent parameters and 12+ sub parameters.
Challenges:
-
Parameters were used differently between departments, which also made us think of a unique parameters for each of them
-
We had to talk to experts to learn how to group many parameters, and also do our own homework to be able to make our work faster, it was interesting and rewarding.
-
We had limited access to the files due to the nature of the data, to figure out deeper how the data might look like.
-
Prioritisation and grouping got slower because we were also unfamiliar of the internal processes.
-
After our efforts we finally could make an Information Architecture of the website that looked a bit like this:
04
Figma prototyping and Usability tests
I translated the user journeys to high-level screens in Figma and built a clickable prototype in Figma. I used reusable Components and used the MUI Figma library modifying them to fit the client's brand identity.
With the created Prototype we did six separate usability tests sessions where each user had to do one search and find a specific piece of information. We prepared a sheet with specific points to test: usability, look & feel, usefulness, relevance of information, gaps found and time of response.
05
Iterations, final user test and handover
We had 12 different usability tests in total and measured the success of the project based on their feedback.
We had a last meeting with the 20+ people involved overall during the project, our stakeholders invited people from other departments to spread the news of the new tool and showcase it to upper management as well. We delivered a 50+ pages document with all processes, challenges, learnings and future plans.
Challenges
-
Limited domain knowledge: we didn’t have access to all data sources because of sensitivity of information
-
At the beginning not knowing 100% how the real data looked like, this put a challenge on the tech side to clean up and learn what types of documents to process
-
Information architecture, relationships and prioritisation due to not being familiar with the company’s internal processes
My takeaways
-
Honed my skills in UI design by creating a search engine from zero, also working closely with developers to build valuable and easy to implement concepts
-
Practiced skills to build user journeys and maps according to current needs and be able to translate to a high fidelity clickable prototype. Used reusable components, libraries and documented all changes for future iterations
I loved this project because of its topic (Health/Medicine) and what it involved (Data analysis) and also prototyping. 😍