.gif)
We were initially tasked with a complicated problem statement that my team had a hard time grasping. Working with AI products was something that was new to all of us, and we were not familiar with the terms our sponsorship team were using to describe their product.
One of the first things we learned in our internship is that IBM practices Enterprise Design Thinking, an iterative process that is referred to as "The Loop" comprised of three stages:
Getting to know users, understanding context,
uncovering needs, getting feedback.
Understanding the team, aligning on intent,
uncovering new insights, planning ahead.
Exploring possibilities, communicating ideas,
prototyping concepts, driving outcomes.
Thanks to the fact that this is a continuous cycle, it means that you can essentially start at any stage of The Loop. Since we were presented the product without a clear user, and without sponsor users to interview, it meant that we were not ready to start the Observe stage. Luckily, following the method of The Loop meant that we were able to start in the Reflect stage.
In order to align with our sponsorship team, we needed to understand the AI tool, and its capabilities to be able to move forward.
The ability for an AI tool to pull
data that is typically hidden in tables and charts.
Makes it possible for a user to communicate
with a machine in the way you would with another person.
Final takeaway: We learned that our AI tool uses Global Table Extraction technology paired with Natural Language Processing, to provide an experience of finding information that is typically hidden in tables and charts by asking the AI tool a question and receiving an answer in a natural language (ex: the way you communicate with Siri or Alexa), which removes the barrier of communication between machines and humans.
After completing the Reflect stage, we felt comfortable moving into the Observe stage of The Loop and continuing forward with conducting some user interviews. Initially, we were told that Data Scientists were potential users for our tool as they typically work with a lot of unstructured data. But after five interviews with Data Scientists, we were uncovering a different story...
So now that we understand our tool, after our first few interviews, we realized that our AI tool is not a great fit for Data Scientists and we wouldn‘t be able to alleviate their pain points. From there, we pivoted and looked for other potential users. We had the opportunity to talk to five students from backgrounds such as:
• Computer Science
• Linguistics
• Statistics
• Business & Economics
While we were able to empathize with them and discovered that they had valid pain points, my team asked what happens to these users once they're done with their studies?
We felt it necessary to make one more pivot. We were noticing a trend in our students who were business majors. They all had similar pain points that our AI tool could help alleviate. Therefore, we interviewed business professionals from various industries. We had a total of seven interviews that spanned over occupations like:
• Entrepreneurs
• Real Estate Agents
• Senior Audit Managers
As a team, we felt like we nailed down our target market, and were able to move on and create a user persona.
In order to reframe our thinking, we completed a Needs Statement exercise, where we put ourselves in our users shoes and asked what is it that they really need.
At this time we were fully aware that our stakeholders have needs as well and that we needed to keep these needs in consideration while designing final outcomes:
The ability to raise awareness of the
innovative work done at IBM.
Creating an engaging and enjoyable experience
in the process of data extraction.
Collect (optional) information from users to target demographics and gauge interest for future iterations of the product.
Creating Hills is an activity in Enterprise Design Thinking that describes something a specific user is enabled to do through our final design.
Keeping our Users and Stakeholders needs in mind, we created lo-fi sketches as a base to what our AI tool would look like and how it would function. Being in-house designers, we also needed to make sure that our tool reflected IBM‘s current visual design and utilized Carbon, IBM‘s current design system.
Prototypes may be showcased when product goes live.
My team was able to conduct usability testing with 5 users, with our biggest takeaways being:
This Enterprise Design Thinking exercise was an opportunity for our team to think big in terms of where we can potentially take the product, and what that would mean for our users.
What we can provide for our users now
What we can provide for our users soon
What we hope to provide for our users later
For this project, I had the opportunity to work on an interdisciplinary team, and was really allowed to work in every aspect of the project.
I allowed my curious nature to take initiative in deep diving into IBM‘s AI tools and learning more about AI design best practices by completing a Team Essentials for AI learning module.
I uncovered a strength of mine which is having the ability to simplify the complex technology and ideas that were involved in this project, and transform them into easy to understand playbacks providing efficient productivity for the team and alignment with our stakeholder team.
Our project was atypical in the sense that none of our members had previous experience in AI, and our stakeholder team didn't have a clear vision on who their product was for, or on expected deliverables. Our initial prototype was based on my lo-fi sketches, and my team trusted me to develop and craft the storytelling for our final playback which we presented to over 60 IBMers where we were overwhelmed with praise and positive feedback for our clearly delivered playback.
.gif)
Being agile and willing to pivot is a great skill. Early on in the internship we were told to fail hard and fail fast. We had no idea how true that advice would be for our team, but it was a great lesson to learn that failure is not necessarily negative, and is part of the learning process.
Being unafraid to disagree with stakeholders, but also being able and ready to back up your thoughts and claims with research. My curious nature lead me to question everything, which greatly helped our team carve the direction for our project, and make recommendations to our stakeholders on how to optimize the tool.
Taking the time with my teammates to do a “show and tell“ of our past projects and our skill strengths at the start of our collaboration would have helped tremendously to ensure easy delegation early on and throughout the project.