Kuebiko -  Empowering Data Analysis and Manipulation

Role: UX/UI Designer
Year: 2023
Highlights: Used by the global data operations team, Handles +150 million POI’s, 200 times faster than its predecessor.

I worked as the lead UX/UI designer at Dataplor, where our mission was to deliver high-quality, human and AI-verified Point of Interest (POI) data to the global marketplace. This journey led us to develop Kuebiko, an internal tool designed to streamline the processing and cleanup of data sourced from various channels. As a result of creating Kuebiko, we've achieved significant milestones, such as elevating data verification rates from 30 to 200 monthly datasets, empowering our users with a more efficient and effective tool for data processing and visualization.

Main screen of Kuebiko Data Manipulation

As the lead designer at Dataplor, I spearheaded the development of Kuebiko in response to the challenges encountered with its predecessor, AdminNext. Initially, our goal with AdminNext was ambitious: to process vast quantities of data directly in the browser and interact with it using SQL. However, after facing stability and speed issues, it became evident that the existing technology fell short of our expectations. So Kuebiko was born, a desktop app capable of processing large amounts of data.

Kuebiko is not a web app like AdminNext was, Kuebiko is a desktop app.

  AdminNext Vs Kuebiko

The Research

Through meticulous user research, including shadowing our users during their daily tasks, and conducting extensive interviews, we gained invaluable insights into the pain points experienced with AdminNext. Surveys revealed that 70% of users were dissatisfied with its performance, while 30% expressed discontent with the user experience.

To address these issues we changed our technology stack and took a step back to look at our users more holistically. Drawing from interviews with AdminNext users, I crafted personas to represent these 2 distinct user types. Meet Jack, our regular user - a data analyst adept at navigating vast datasets but lacking in advanced technical skills. For Jack, simplicity and ease of use were paramount. In contrast, Maria, our advanced user persona, represents data managers and engineers who require robust tools for complex data manipulation and analysis.

Jack User Persona

Maria User Persona

Pain Points

In addition to the technical enhancements made to Kuebiko, we recognized the importance of equipping users like Jack with the necessary knowledge and skills to leverage the platform effectively. To address this, I introduced improved user training initiatives tailored specifically to data analysts’ level of technical expertise. These initiatives include the development of user-friendly tutorials and interactive guides focused on essential features and functionalities relevant to Jack's role as a data analyst.

Additionally, we offer hands-on workshops and virtual training sessions to ensure data analysts understand how to navigate and utilize Kuebiko effectively for data analysis tasks. By investing in data analysts’ training and skill development, we aim to enhance user confidence and proficiency, ultimately maximizing the value derived from Kuebiko's capabilities.

Jack's user journey

The Design Goal

Kuebiko emerged as a response to the evolving needs of our users, requiring a deep understanding of both technology and user requirements. My role as lead designer demanded an intricate balance between technical complexity and user-friendly design. To address this, I focused on three key pillars: simplicity, user freedom, and progressive disclosure. By prioritizing simplicity in design, I ensured that users, regardless of technical expertise, could navigate Kuebiko effortlessly. User freedom was paramount, allowing users to interact with data in a way that suited their preferences and workflows. Additionally, progressive disclosure ensured that users were presented with information and features in a manner that matched their level of expertise, fostering a seamless user experience. Collaboration with the engineering team was crucial in translating these principles into a feasible and powerful platform. Through meticulous documentation and comprehensive training initiatives, I ensured that users could leverage Kuebiko's capabilities effectively.

A list view of kuebiko deliveries

Iterations

Kuebiko's journey of iterations began with version 1, which included a dropdown with all the filters. However, this version struggled with menu hierarchy, hindering user navigation. Version 2 aimed for modularity but failed to adapt to all screen sizes, lacking horizontal real estate. With version 3, Kuebiko became significantly faster but still resembled AdminNext too closely, retaining spacing issues. Version 4 introduced a more compact design and a dropdown that accommodated for more options, aligning better with our business objectives and paving the way for future additions.

The different version of AdminNext and Kuebiko

The capabilities of Kuebiko extend far beyond its primary interface for live data interaction. In addition to facilitating real-time data manipulation, Kuebiko empowers users by providing comprehensive access to our entire dataset inventory and detailed analyses for each dataset. This includes the ability to list all datasets available and delve into detailed analyses, forming a complex data hierarchy. I meticulously composed and tested this hierarchical structure through diverse iterations of menus and pathways, ensuring intuitive navigation and efficient achievement of user goals.
A main menu > breadcrumbs > tabs

The content hierarchy of Kuebiko

Slide 1: Canvan view for task queue, Slide 2:  JSON file display for fill rates, Slide 3: Deliveries information about Exports

Next Steps

The next step for Kuebiko is to include Validators, a new type of user. With fast-paced tasks, this new section will enable the data team to incorporate diverse methods of data validation in a fraction of the time.

In conclusion, Kuebiko stands as a testament to our commitment to continuous improvement. Our iterative process, fueled by rigorous qualitative smoke testing and user feedback, drives the evolution of Kuebiko's features to better serve our diverse user base. By incorporating user insights and introducing new functionalities, we not only enhance Kuebiko's functionality but also ensure its reliability and usability in real-world data analysis tasks.

A task of the Validators new persona

Other Projects

UX/UI 

Dolores - Redefining Business Data Verification with a Chat Bot

UX/UI 

dataPlor Explorers - Transforming Business Data in LATAM