AWS Datazone
A data management service that simplifies data sharing and governance across organizations, enabling teams to catalog, discover, share, and govern data stored across AWS, on-premises, and third-party sources.
01/ Overview
My Role
UX Design Lead
Year
2022 - 2024
I played a pivotal role in the cross-functional team that successfully launched a new service. Collaborating closely with Product Managers, API, Front-end, and Back-end Engineers, SDMs, and Solution Architects, I spearheaded the design and development of its data cataloging and discovery features from the ground up.
These features launched in October 2023 and quickly gained traction, reaching thousands of active users within the first month. Leveraging insights from an initial limited release to over 14,000 users in November 2022, I informed critical decisions that led to a highly successful general launch in October 2023, scaling the service to over 100,000 users and securing adoption by more than 15 major businesses.
Timeline & Process
Key Deliverables
-
Govern Data
By applying metadata forms and glossary terms to assets, data producers can improve data discoverability.
-
Metadata
Templates & categorization that standardize the collection and management of metadata to ensure consistency, accuracy, and usability.
-
Data Discover
Data consumers can easily find data assets by applying relevant filters,and view data asset details.
02/ Research & Analysis
Interviewing executives, Solutions Architects, and users gave me insights into the product vision, technical aspects, and user expectations. This research helped pinpoint the problems and key user pain points we need to solve.
Objective
1- Define the users
2- Identify their key pain points
3- User Journey with Datazone
Methodology
Shadowing
Spent time with users to understand their workflows to gain first hand experience of their pain points
User Interviews
Conducted 20+ in-depth user interviews to further understand their needs & requirements
Data from Solution Architects
Leveraged their direct client interactions to collect insights to identify key issues
Who are our users?
Data Producers / Stewards
Experienced in data management and governance, including both structured and unstructured data.
Pain Points
Data Silos and Fragmented Access
Businesses operate with data scattered across numerous independent databases. This necessitates accessing each database separately to retrieve and manage information.
Time-Consuming Processes
The need to connect to and query multiple disparate databases for analysis or reporting slows down critical business processes and decision-making.
Inefficient Data Management
Dealing with data in isolated systems leads to redundant efforts and a lack of a unified view. Routine data management tasks become cumbersome and require significantly more time and resources.
Scalability Challenges
As data volumes grow and the number of data sources increases, this fragmented approach becomes increasingly difficult and costly to maintain and scale effectively.
Data Producers / Stewards
Discover and access high-quality data to generate insights, then collaborate with colleagues to validate findings & share them, all with a less technical approach.
Time Spent on Data Search
Analysts waste valuable time navigating disparate systems and searching for the specific data they need for their analysis.
Difficulty in Data Discovery
Finding relevant data assets is a significant hurdle in large organizations with data spread across numerous departments.
Identifying Data Owners
Determining the correct data owners within the organization adds another layer of complexity and time to the data acquisition process.
Access Request Delays
The necessity of requesting and obtaining permission to access data from various owners introduces further delays in the analytics workflow.
03/ Design & Iterations
From the preview launch(limited users), I gathered crucial feedback, specifically focusing on usability, user expectations, IA and navigation. The learnings from this initial phase directly informed necessary refinements and optimizations to the user experience. This iterative approach, prioritizing user feedback early on, minimized risk and ultimately facilitated successful widespread adoption
High Level System Diagram
04/ Preview Launch
Datazone was preview-launched in November 2022 to internal users and select clients. Learnings & feedback from this preview was used to refine and improve the design ahead of the full launch in October 2023.
05/ Feedback & Learning
1. Observational Research (Shadowing)
To gain firsthand understanding of user behaviors and identify potential pain points within their natural environment.Shadowed data producers and consumers actively using the service.Observed their workflows for data governance, data discovery, and data utilization.
Following the preview launch, I spent three months collecting usability data to identify areas for improvement.
2. User Interview
To comprehensively understand customer needs, uncover areas for functional improvement, and gather qualitative feedback on their experiences.Shadowed data producers and consumers actively using the service.Conducted over 200 in-depth user interviews.
3. Collaboration with Solution Architects
To leverage their direct client interactions and insights to identify key issues and challenges faced by users in real-world scenarios.Collaborated with 10+ solution architects who were handling more than 50 clients.
Addressing the UX Issues
The preview launch surfaced significant usability and UX hurdles, revealing key discrepancies between our assumptions and actual user behavior. I categorized those issues in three categories.
1- Structural Misalignments
2- Expectation Gaps
3- System Complexity & User Support
High level UX issues identified during preview phase
06/ The Launch 🎉
After a year of multiple iterations, intense discussions and usability evaluations, the service was finally launched.
07/ Challenges
Complexity of the Service
Technical omplexity of our service and how it connects with other in the vast world of AWS.
Lack of Competitive Products
We were the first one to launch something like this, so it was hard to find other competitor who were doing similar designs.
Unclear Vision
I was only given a one-page explanation, and I facilitated the initial sketching and definition of the service.
Limited Time and Resources
With limited time and resources for the preview launch, I had to collaborate closely with the tech team to ensure that my designs had a high impact while minimizing tech effort.