CASE STUDY
Combining 15 Dataset-Specific Apps in a Single Analytics Platform
Client
Nielsen IQ, a data and analytics firm covering more than 90% of the world's population with operations in 90+ countries.
Solution
Create a a dashboard experience which allows users to curate the data they are most interested in across datasets.
Once launched, the product went from 0 to 50k users in 12 months.
Problem
How might we create an unified data analysis experience for 15 distinct datasets previously siloed in single-use apps?
My Role
Collaborated with another designer, Pedro Iracheta, and a design manager, Tony Lopez.
Delivery Overview
Reflecting the user's mental model in the interface
Previously, users had to construct high-level data overviews on their own, or ask an expert to do it. Now key business data is collated across datasets front-and-center.
Reduced time-to-task and cognitive load
Reports can now run automatically, and users can set automatically updating KPIs and alert queries.
Discovery
The business requirement
High-level user needs
I was tasked with designing the dashboard for the new product: the central terminal between users, the numerous datasets they wanted to collate, and the most pressing high-level information, which would they would curate themselves.
Before I joined the project, UX Researcher Don Hiles conducted thorough interviews and distilled his findings:
​​
-
The main users worked either for Manufacturers or Retailers. They wanted to increase sales, grow distribution, and understand consumer behavior in order to drive strategy.
-
NIQ's 15 different datasets were for the most part not cross-functional. If a user wanted a particular piece of data, they had to access the application specific to that dataset.
-
Users were improvising their own custom tools to pull data sources together
-
Users would switch between 4-7 different reports or tools in order to identify answers to business questions
Iteration
Designing the MVP
In parallel to discussions with stakeholders regarding business needs, we iterated our way to a design that we were ready to test with users in order to get their feedback.
​
BUSINESS NEEDS
-
Competitors had outpaced NIQ and were able to offer dashboards which made it easy to find relevant data; we needed to do the same
-
​Instead of segmenting the page by dataset, the overarching strategy for this new platform highlighted cross-functional dataset navigation; therefore, the dashboard needed to be the launchpad for this strategic functionality
DASHBOARD-SPECIFIC USER NEEDS
-
Surface high-level insights: Power-users were not going to user the dashboard, because their business processes suffered from any degree of simplification; our dashboard needed to be digestible by everyone from junior analysts to C-suite executives.
-
Make everything customizable: The wide variance in user business context (from manufacturer to retailer to Wall Street and beyond) required a balance of customizable KPIs and alerts as well as system-recommended content
-
Make KPIs dynamic: Our users are specifically interested in performance *as compared* to competitors, including both specific companies and industries as a whole; our KPIs needed to allow dynamic benchmarking
-
Alert functionality – Users are juggling numerous spinning plates at a time, so they should be able to configure alerts for specific events and outcomes
Information hierarchy / data source variations
In each of these iterations you will notice variations in the relationship between A) Dataset B) KPIs and C) Reports. The configuration of these elements is incredibly impactful for the dashboard experience as a whole: some users will be in charge of many products and markets, while others will have a narrower focus. We wanted to make things simple and straightforward for everyone.
Further Iteration and Testing
Pushing the KPIs further
We had a sync with our PM and our engineering team, and decided that time constraints required us to focus on KPI and alert updates after the MVP launch, so that's what we did, focusing directly on improving based on our usability tests.
KPI update 2.0
We began with the easiest fix as a holdover until we could allocate the resources to update KPIs again to be even more comprehensive. We added a karat to the card so users could more easily read and scan the individual datapoints constituting the KPI.
KPI update 3.0
In addition to adjusting the type sizes in order to be more readable, we also included spark charts in order to help users visualize their KPIs with more context.
Feedback
"
We weren't pretty impressed with [this product.] We were really, really impressed with it. I called our client lead and asked why we weren't on it now.
Senior Manager | Multinational Agriculture Corporation
"
I really like it. Beautifully visualized and clear. If I can do it, anyone can do it; from senior leadership to working group. Simple and clean, enables analytics. I just want to see more of it.
Chief Digital and Marketing Officer | Multinational Cosmetics Company
I had the pleasure of working with Alex for close to a year when our company hired his employer to help with UX designs. He’s a very talented UX designer who is great at listening to user needs and product vision to design effective product solutions. Alex delivers and is very reliable, not to mention he’s collaborative and pleasant to work with. I very much enjoyed working with Alex.
"
Karolina Roszel, Vice President, Product Leadership | Nielsen IQ
More case studies
Modernizing Revamping a Complex Government Services Platform
Redesigned core functionality for a platform that serves hundreds of municipalities across the country.
Bringing Capital Group's ESG Mini-Site to Life
I worked with Be Grizzlee to redesign a suite of pages for the financial services company.
Creating a Core Service for Nielsen IQ, an S&P 500 Consumer Intelligence Company
Redesigned a key ecosystem within a new SAAS offering, from ideation to delivery.
My Writing
My background as a journalist informs my designs: it helps me think through tough questions and deduce actionable conclusions.