Knowledge-base

Problem

Amazon Web Services (AWS) encountered a significant challenge managing data across different internal processes, each catering to distinct stakeholders. The initial approach involved disparate systems and methodologies, leading to inefficiencies, data inconsistencies, and collaboration challenges.

‘Decisions are made on a case-by-case basis, as there are a number of data points to consider, so we rely on business analysts to provide advice on the next steps.’

MANAGMENT LEAD

Requires to see high level data in a broader sense in order to make informed decisions

‘I have a spreadsheet that I update every week. I'll email it to my manager (she can make changes - he'll let me know in Slack), and he'll integrate it with my colleagues' analysis... not sure how they do their spreadsheets.’

BUSSINESS ANALYST

Creates a data management nightmare, relying on manual processes that are error prone

Microproblems⁷ ↴ ↴ ↴

↩ See All ↩

Data management systems often struggle to effectively integrate forms, tables, and graphical reports, leading to complex and isolated information management. Solution:"The Data Blueprint" offers a systematic approach to design, focusing on user experience and integrating key components for coherent navigation. It includes steps like filtering/sorting data, detailed views, data modification, and collaboration, each with specific UX considerations. Insight:This approach reflects common practices in various applications, emphasizing the importance of understanding and utilizing established UX principles in designing data management systems.

Users performed repetitive tasks that required extensive navigation, leading to system inefficiencies. Solution: A 'Quick Tasks' option was introduced on the homepage, designed to streamline and simplify the process for users, allowing faster access to frequent tasks. Insight: This change resulted in a significant increase in productivity, with users logging information 200% more than before, indicating the effectiveness of user-centric modifications in enhancing system usability and efficiency.

Users struggled to identify individual data points grouped based on business logic, making it challenging to locate specific information. Solution: The implementation of a tagging system allowed users to categorize and find data points more efficiently based on related tags. Insight:Although the solution improved data point accessibility, it required significant administrative effort to ensure the tags were well-defined and relevant, highlighting the balance between system enhancements and operational workload.

Users faced limitations in choosing options for various fields due to dependencies between them. Changes in primary fields necessitated updates in related secondary fields. Solution: We developed a 'field dependency' feature, which dynamically adjusts available options in secondary fields based on primary field selections. Insight: This solution provided users with clear guidelines for data entry, significantly enhancing data quality and consistency within the system.

In a data gathering application where accuracy is crucial, users struggled with complex information that was lengthy, difficult to connect, and challenging to search for. Solution:We integrated an Artificial Intelligence framework, visually distinct from the rest of the application, to clearly indicate AI-generated versus human-entered information. Insight: This approach helps users distinguish between AI and human inputs, enhancing trust and clarity in the veracity of the data presented.

In a data gathering application, ensuring the veracity of information at specific times and facilitating dynamic collaboration were key challenges. Solution:A notification system was developed, enabling users to stay updated on data changes and collaborate effectively within the system. Insight:The transition to enhanced collaboration highlighted the importance of providing links for discussion. Moving forward, we emphasize direct communication in the system, fostering transparency and efficiency in collaborative efforts.

An application initially designed for a small, regular user base suddenly needed to accommodate thousands of sporadic users, requiring a new approach to provide varying levels of documentation and guidance. Solution: We developed a tiered documentation strategy, addressing everything from field-specific values to the application's overall purpose, catering to the diverse needs of the expanded user base. Insight:We realized that providing high-level context for each object was crucial. Starting with a comprehensive wiki and establishing a point of contact emerged as an effective Plan B to support our diverse and growing user community.

Mockup of Final Product

Proposal & Process

It was proposed to create a tool with multiple modules serving different stakeholder groups. Each module caters to a specific topic while providing data collection, data maintenance through collaboration, and report generation. These are the steps taken to achieve the above:

1. Examining stakeholders' expectations and requirements for each module following Amazon's "working backwards" approach.
2. Through user interviews, identified the flaws and benefits of existing data collection, collaborative maintenance, and report generation processes previously used by users.
3. Developed concepts for navigation of interconnected modules, as well as prototypes for data collection. These drove discussions regarding user needs and usability issues. The results of these discussions were used to create user-centered designs. The designs were implemented and tested by users, leading to improved user experience.
4. Organizing multiple feedback sessions with module stakeholders and future users.
5. Adapted AWS's design system to the visual representation.
6. Interviewed stakeholders and conducted usability studies to gather insights into prototypes.
7. Iteratively Incorporated feedback to refine the user experience.
8. Ensured seamless interconnection of modules according to changing requirements.
9. Wrote user stories for developers based on growing changes and additions
10. Research the usage
11. Iterate

Bringing developers into design sessions was crucial for feasibility and understanding MVP stages and evolution. Most of the above activities were repeated over several Sprints.

Initial workshops to understand project MVP

Result

Ultimately, the newly developed internal tool transformed how AWS handled data. It empowers the management team to make decisions using up-to-date and trustworthy information. The interconnected modules facilitate smooth task transitions, enhancing collaboration. Unlike the previous method with information silos, users can now gather, maintain, and generate reports collaboratively without barriers.

Initial designs for data gathering processes

Impact

Due to NDA restrictions, exact figures cannot be shared, however, the improved efficiency of collaboration has resulted in significant cost savings and increased productivity allowed for the following:

• Data centralized management ensures uniformity and consistency in storage and retrieval, reducing the risk of data discrepancies.
• Validation and quality control were implemented through a centralized application, which minimized the chances of incorrect or incomplete data entry.
• Enhanced security due to centralized access controls and data encryption.
• Effortless reporting and analytics were facilitated through a unified and up-to-date database, which enabled meaningful insights and informed decisions based on reliable data.
• Cost Efficiency was achieved by consolidating data into a single application, reducing redundancy and streamlining operational processes.

Challenges

There were many challenges in this project, including:

• It was challenging to reconcile the diverse needs and preferences of different stakeholders. Iterative design required a delicate balance between design and functionality. Working closely with stakeholders and users throughout the process was crucial to overcoming this challenge.
• The transition for users who were accustomed to the previous module-specific workflows was challenging, as they had their own processes and ways of working.
• As the users handled sensitive information, there was limited access to the production site in addition to limitations on the discussions of data-specific topics.

Role

This project required me to work overtime in 3 big hats:

Senior UX Consultant

My role involved orchestrating the unification strategy, collaborating with diverse stakeholders, conducting thorough user research for each module, and overseeing the iterative design process. Bridging the gap between distinct stakeholder requirements and aligning them with AWS's design system was a pivotal aspect of my role.

Account Manager

Played a crucial role in nurturing strong client relationships, coordinating seamless project delivery with internal teams, and ensuring effective communication channels between AWS and Everest Engineering. I was responsible for managing client expectations, overseeing project scopes, and actively seeking account growth opportunities. My role involved balancing client needs, internal team dynamics, and project delivery, which contributed significantly to client satisfaction.

Team leader

Leading a software development project involving six software developers and a business analyst. Prioritized unblocking team members by addressing their difficulties as soon as possible. My goal was to assist each team member in developing their career through tailored guidance, identifying learning opportunities, and encouraging skill development. Over the course of the project, I acted as the primary point of contact for the team in operational matters, maintaining effective communication channels between the team and stakeholders, and maintaining a positive team culture.

Learnings

From this project I got three main takeaways:

• Consulting as a skillset: professional consulting skills played a pivotal role in facilitating meaningful discussions with stakeholders, clients and existing research. Guiding the client through iterative design processes, user interviews, and feedback sessions ensured their expectations were met. It also enabled them to make informed decisions aligned with the project's goals. Ultimately, the success of adapting to stakeholder needs hinged on the ability to effectively consult, translate, and guide the client through the complex landscape of design and development.
• Value of team culture: despite the initial challenges associated with remote working and contractor-based teams, this project highlighted the importance of developing a strong team culture. As a result of improving team cohesion, organizing regular virtual meetings, and encouraging open communication, we mitigated early struggles, enhancing collaboration and cohesion among the team. Creating a positive team culture not only improved work dynamics, but also contributed to increased productivity.
• The Amazon way: two concepts in ways of working stand out: "work backwards" and "mechanisms." Working backwards means starting with a desired goal and working backwards to figure out the steps to get there allowing yourself to change while discovering more about what you are building. Using the mechanisms approach, you break complex problems down into simpler parts and find solutions from there.