Data Mapping, Cleansing, and Alignment

Introduction

A major challenge for researchers at MSK is that valuable data points are often buried in unstructured text within doctors’ notes. As a result, many manually curate data into spreadsheets for their research projects. Over the past few years, my UX team and I have developed solutions to capture this curation process and transform it into a reusable data asset.

We conducted extensive user research, interviewing researchers and building prototypes to refine the experience. Our efforts led to the improvement of MSK Extract, a tool now available to all MSK researchers. MSK Extract enables researchers to:

• Create a curation database without needing SQL or ER diagram knowledge.
• Import spreadsheet data into a custom database
• Map data elements to MSK’s standard terminology.
• Share curated data with MSK for reuse

Our intuitive design has made adoption seamless, with some researchers onboarding without needing training materials. Over the past three years, more than 300 databases have been created using MSK Extract, streamlining research workflows and improving data accessibility.

In user interviews while reviewing MSK Extract, most researcher were purplxed at the opening screen. 

While the team wanted to make the app friendly with a personalized greeting, the UI wasn’t clear about what the users should do. 

We see several cards all with the same weight, items that are clickable are not easily identified, and the copy is verbose. 

Users we’re also confused by the “Shop” menu link. “What am I going to buy here?” was a common comment.

Wizard, a native Apple app that I have used for well over a decade, makes finding patterns in data really fun. I always have admired the onboarding process – because there are ALOT of moving parts in the application.

One of the best things is the introduction journey; where the creators have given yo a simple data set to play with to explore features.

We call this a “Warm Welcome” and I knew that this was the proper way to re-invent the home page on MSK Extract. 

The primary stakeholder introduced my UX team and me to the project while our accounts were being provisioned. Originally designed to meet her specific needs, MSK Extract was now our responsibility to scale for all researchers at MSK.

After interviewing every user who logged in the previous year, two key insights emerged:

1. 25 out of 32 users mistook the onboarding conversation for training.
2. All users had abandoned projects in their cart.

These conversations also uncovered a “top ten” list of frustrations and desired features, providing invaluable insights that shaped the product’s evolution and ensured it better met researchers needs.

During research readouts, I identified recurring but unspoken frustrations:

1. Corrupted data
2. Data files too large to process
3. Lost or unbacked-up data
4. Data localized on individual workstations
5. Forked datasets caused by duplicate updates
6. Most users came to Extract expecting to find answers

These issues stemmed from the traditional pipeline structures common in research organizations. I proposed abandoning the pipeline model in favor of a cloud-based, project-oriented structure. By retiring the ad-hoc feature-driven approach and emphasizing componentization, we could expose core functionalities through APIs. This concept was well-received by the architecture team, initiating our migration to the cloud.

The most critical research finding was that users expected Extract to provide answers, not just tools. Many didn’t realize they needed to bring their own data, assuming they could simply “extract” what they needed from a central database.

This insight was eye-opening for the team, fundamentally shifting how I approached user education and onboarding.

The UX team and I created hundreds of designs, and here’s a small selection of key highlights. I won’t dive into all of them, as they are already well-documented.

I require my junior designers to produce detailed documentation—not as a formality, but as a critical safeguard. Over nearly two decades of wireframing, this practice has saved us countless times.

Feel free to explore—thousands of hours of work are now at your fingertips.

With UX best-practice wireframes handed off to the dev team to improve usability, I could focus on the bigger picture—scaling Extract for every user.

The UX team at MSK follows a rigorous Design Thinking process to ensure not just that the work gets done, but that it’s done in the best possible way. Through iterative testing—from interviews to high-fidelity prototypes—we uncover critical insights that drive product development.

Key findings from our research on Extract included:

• Different research silos used different terms for the same operation
• Researchers wanted to use MSK Extract but didn’t fully understand what it was
• Many abandoned projects because they didn’t know how to prepare their data
• Users were confused by the concept of a “Shop” in a data application

With Extract gaining attention across MSK, I had the opportunity to grow my team—and Lina and Mike were standouts.

Lina was my right hand, working alongside me to conduct research and uncover key insights. Mike, on the other hand, was “no hands”—we’d discuss a direction, align on the research, and he’d take off, always returning with exceptional designs.

High-Fidelity Prototyping Beyond Figma

While our wireframes are created in Figma using the robust, proprietary MSK Design System, we quickly reach its limits – especially when dealing with complex, data-driven interactions. Given the exponential variability of research data, I build high-fidelity prototypes that follow an MVC or Observer Design Pattern. These prototypes, though quickly assembled, allow for more realistic user testing and validation.

One day on a conference call, I spotted a familiar face—my longtime friend from my software engineering days, Clacius Clecidor. It was pure serendipity, and I immediately worked to bring him onto my project.

I needed an expert full-stack engineer, and Clay had been a key member of my “Dream Team” for over a decade.

Below’s a link to a high-fidelity prototype I created for testing. While some elements may no longer function, it offers insight into our design process and approach to solving complex UX challenges.

Three years later …

We have gone through several phases: Import, Decoupling, Move to Cloud, Data Governance, and most recently, integration with the cBioPortal

We have continued to conduct extensive user research, interviewing researchers and building prototypes to refine the experience. Our efforts led to the development of a tool now available to all MSK researchers.

MSK Extract enables researchers to:

• Create a curation database without needing SQL or ER diagram knowledge.
• Import spreadsheet data into a custom database.
• Map data elements to MSK’s Standard Library of Terminology Concepts (over 1500).
• Share curated data with MSK for reuse.

Our intuitive design has made adoption seamless, with some researchers onboarding without needing training materials. Over the past three years, more than 300 databases have been created using MSK Extract, streamlining research workflows and improving data accessibility.

Reflecting on three years of work with MSK Extract is both rewarding and exhausting. It has been a journey of learning, innovation, and collaboration, embodying the principles of design thinking and pushing the boundaries of what’s possible in oncology research.

Below is a gallery of remanence – behind the scenes files that show the detail of the work. Included are screen shots of Miro Boards, Confluence pages, and Figma workspaces. 

My Role in Product Strategy