IBM Blocks
User Research, UX

IBM Blocks is a data analytics platform that allows data analysts to build analytic structures, gain insights, and easily collaborate  through simple, modular forms, just like building blocks.


Blocks was a 6 week incubator project focused heavily on understanding the data science work process, the people involved, and uncovering the most immediate needs/pain points in the process.


User Research

UX Design

Collaboration with

Wonhee Baek- Research
Sean Gardener- UX
Amy Le
- UX
Pablo Domowicz- Front End Dev

Ann Novelli- Visual

Grainne Smith- Visual

User Interview

We conducted interviews with both internal IBMers and non IBMers involved in the data science process to understand the domain, the processes involved, and answers assumptions that

the team had around the data science process. We conducted in person interviews, phone interviews, and usability tests.

User Journey

Primary User
Business Analyst "The reporter"

Pain points:

  • Analysts have to rely completely on developers to get any of this process done

  • Analysts have trouble understanding technical jargon

  • Analysts rarely have the specific information they need to make confident decisions without relying heavily on intuition

  • Analysts don’t have ownership of the process so it’s difficult to justify outcomes

Secondary User
Data Scientist "The builder"

Pain points:

  • It is difficult to keep track of and merge all the different datasets in a single place

  • Segregated tools and lack of context delay the progress of projects

  • Developers constantly have to build new tools to answer similar questions

Data Science Ecosystem

IBM Blocks is an ecosystem where everyone is here, and everything is here. It’s one place that you never have to leave, because everything you need is provided. You can reuse your materials, so you don’t have to go through the tedious process

of rebuilding everything from scratch, each time around. IBM Blocks is constantly evolving with you, learning your business preferences, taking in new data, and training its analysis model so it can provide the accurate insights you want.


A  Data Analyst can build big data applications to make business decisions 80% on their own while the system provides suggestions and predictions.


A Data Scientist/Developer can use a single environment with preconfigured offering to assist in coding applications participating in only 20% of the process.


A Business analyst can generate insights from dynamic programs that allow for continued use without needing a Data Scientist/Developer.

Post Incubator

After the 6 week incubator project, Blocks, was complete, I eventually joined Analytics Platform. There, I continued working on this vision on the Next Gen Workbench and the Data Science Experience “Canvas".

Lo-fi, Mid-fi Wireframes