Research Case Study
Information provided in this case study is limited due to non-disclosure agreement. For access to full case study, email me.
Team
UX Researcher - Leah Heifetz
Director of Product Design
Tools
Trello
Respondent.io
Figjam
Zoom
Otter.ai
Duration
5.5 weeks
Skills
Generative Research
User Interviews & Surveys
Affinity Mapping
Data Synthesis
Analytical Thinking
Storytelling & Presentation
Overview
The Context
A data solution company wanted to learn about a specific user group of their product: data buyers. They approached me to be part of the 5 week sprint.
My Role & Team Collaboration
UX Researcher alongside the Director of Product Design, Founder, and intern.
Weekly sprints and stand-ups to review completed tasks, tasks to be done, and clarify unanswered questions.
Tickets on a Kanban board to prioritize tasks. dir. Product Design led the project direction and managed tickets for the team, while I created and delegated various research tasks for the intern.
click to zoom
Research Goals
Problem Statement
The CEO and team wanted to speak with data buyers. I helped them identify the problem:
“We want to better understand how users make decisions on buying datasets and how they interact with datasets prior to and after purchase, in order to validate and improve the product.”
Research Goals
Discover data buyer’s current processes/decision-making about consuming data, and how they feel about the overall experience.
Learn about users current pain points, frustrations, and barriers about the data buying process.
Uncover the current tools data buyers are using for data consumption and processing into their systems, their experience with them, and how they would improve the tools.
Validate or invalidate design decisions based on user needs.
Discover unidentified industries that purchase datasets and evaluate their consumption habits compared to other industries.
Methodology
What is the desired outcome of the research?
Work backwards to determine which research method made the most sense.
Define the outcomes in order to define the questions and the best research method.
Consider restraints. I knew I needed a research method with a quick synthesis turn-around time.
Enter: Screener & User Interviews
To make sure I was talking to the right people and getting the right information I set up a screener through Respondent to weed out any bad candidates. I formulated questions about data consumption habits as well as data security needs. Candidates were screened based on job title, amount of data consumed, their interest in data security, and data sharing needs
15 interviews conducted over Zoom
International panel from varying industries
Interviews lasted 1 hour
Recruitment
“Seeking data buyers and related professionals who have recently purchased large datasets.”
I recruited C-level professionals (think: Chief Data Officer) with decision-making abilities regarding tools, products, and processes for data consumption because they would provide the most relevant answers to our questions and could potentially reveal barriers to adopting the product.
Finance and healthcare were identified as industries that consume large amounts of data, have a need for security, and data lineage. I asked 11 screener questions to weed out poor candidates that didn’t match the profile we wanted.
Sample Screener Questions:
“Please describe an instance involving the purchase of a large dataset. What was your role in this transaction?”
“How do you ensure a dataset is secure and not tampered with throughout its life cycle prior to and during consumption?
User Interviews
I used interviews to gather qualitative, attitudinal data. I wanted to hear users talk about the why and how of data consumption. Since this was discovery research, I crafted broad questions, focused on the tasks users need to complete, and made no mention of the product until after the interview if participants showed interest.
Interviewing a participant via Zoom
Interview Questions
click to zoom
Analysis & Synthesis
After each session I wrote down the top 5 takeaways from the interview in order to start spotting patterns among interviewees. I then downloaded the recording and transcribed each recording in order to use for note taking. I used the recordings as reference points during the synthesis to go back and listen to and pull out more points for analysis.
From here it made sense to build an empathy map for each participant in order to understand the motivations behind their behaviors. I looked at their behaviors/attitudes, needs/goals, frustrations, and telling quotes.
I then used the empathy maps to create an affinity map to visualize commonalities and interact with the data. Emerging themes included pain points, desired features, industries, and data types.
The affinity map served as an outline to create a written report for the stakeholder.
click to zoom
Outputs & Deliverables
click to zoom
The client requested a written report and plugging data insights into their spreadsheet tracking system. I still opted to create empathy maps and an affinity map to identify patterns across the data and categorize findings. They also served as a building block for the written report.
Taking into consideration the client’s needs, in a final team meeting I briefly presented the affinity map and took a deep-dive into the written report and Excel spreadsheet. I highlighted the biggest takeaways - the data buyers profile, user pain points, and feature opportunities - and gave recommendations for next steps.
Impact & Next Steps
Information provided in this section is limited due to non-disclosure agreement. For access to full report, email me.
From output to impact
Validate design decisions. The Product team used insights to support product features. The consistent mention of the same pain point across interviewees confirmed the team’s hypothesis and was used to validate design decisions.
Understand competitors. Through market research, I identified another significant data marketplace. While not a direct competitor, this discovery was crucial for the client to understand how others are utilizing datasets. Additionally, I emphasized the importance of data security regulations, which the company was still in the process of fully understanding.
What’s next?
Build new features discovered from research once the product gets to beta launch
interview the next identified focus group based on recommendations from my final report
Concluding Thoughts
What went well?
Team dynamics. The team worked very well together as we all had a similar working style of being results-driven, agile, and collaborative.
Prep work. The interviews ran smoothly from a technical and administration standpoint. I performed several checks ahead of time to make sure there would be no technical problems, and communicated with the participants if there was a scheduling issue.
What didn’t go well?
Subject matter. I initially was caught off guard if an interviewer asked me to elaborate on a question from the script. Given that I am not a subject matter expert on datasets, it took me a few rounds to feel comfortable enough with the industry jargon to go off script.
In the future I would make sure to familiarize myself more with the research content so that I am able to confidently speak to it with the participant. The more comfortable I am, the more comfortable they will be as well to share and elaborate in their answers.
Challenges encountered
Note taking. I did not have anyone to take notes during the interview so I spent a lot of time watching the recording to take notes. I would have liked to fully focus on the interview and not worry about jotting down quick notes throughout each session.
Product complexity. The market for the product was challenging to comprehend, making it difficult to understand how the product functions. Although I didn’t need to pitch the product with participants, a deeper understanding of it would have enabled me to better grasp how it could meet their needs. This, in turn, would have allowed me to ask more insightful questions based on their feedback during the interview.
Room for improvement
Fewer interviews. I interviewed 15 people for this project per the client’s request. Next time I would save time and resources and put a cap at 7 or 8 interviews, because at a certain point they all said the same thing. Also, the more we interviewed, the less the participants fit our candidate profile so their answers did not provide as useful information.
Better facilitation. Finally, I would be more succinct in guiding the interview conversation to stay on topic and be conscientious of time.