Talent PRO is a Talent Management & Workforce Analytics tool that is powered by Machine Learning & AI and was one of the biggest projects I worked on with PiSquare, a Data Science company that used ML Models & AI to make workflows & modules (name of the tool is different in real). Talent PRO is a comprehensive tool that visualizes data across different facets of a Workforce, for example, Recruitment, Performance, et cetera.
Background
The need for a ‘new’ Talent Analytics tool emerged as we onboarded a new client, a massive biopharmaceutical company based in India that had the requirement for a tool for internal operations & workforce management. We already had something built before in that category however it was not nearly good enough for the client’s requirements. Hence, we began rethinking the entire user experience. We picked up several elements & processes from the structure we already had with us but now we needed new ideas to contain the vast amounts of data we’d have on our hands, ranging from different facets of a multi-national organization & time periods.
Problem Statement
Just like with every product, I began with understanding the problem space a little better. Basically, the need was…
…for a platform that could give a holistic yet bifurcated view of the key metrics related to Talent & Workforce management for any organization. It also needed to have the capability to see forecasts & predicted data for the upcoming quarters which would give the user ample time & resources to identify & plan out strategies.
Since we had a very tight timeline to execute this, we had to follow a very lean & agile methodology to go about our objectives.
What exactly is Lean UX?
Lean User Experience (UX) Design is a user-centered design process that embraces Lean and Agile design methodologies to reduce waste and build products centered around the users. Lean user experience design relies on a collaborative approach and rapid experimentation/prototyping, to get user feedback by exposing a minimum viable product (MVP) to users as early as possible.
It has three stages - think, make & check.
Think - Brainstorm possible areas for improvement based on feedback or problem statement. Do some secondary research, competitor comparisons, and observe how users are using their product. Based on generated insights, outline potential solutions.
Make - Sketch out low-fidelity wireframes or high-fidelity prototypes as per the requirements to get a better, concrete understanding of the solution.
Check - Product teams test the new feature, using tools like UX surveys and A/B testing, to figure out whether their hypothesis was correct. If customers respond well to the new feature, it becomes part of the new design. If it doesn’t improve the customer experience, they return to the Think phase and try something new.
Thinking
Identifying End-Users
We first identified the profile of users who would be using this tool to track & analyze KPIs. It mostly constituted of Managers, Department Heads & Organization Heads who needed a macro-level view of the organization & its departments.
Conducting a User Interview
To understand the end users’ needs & pain points better we decided to interview a manager from the organization we were designing this product for. The identity of the manager is confidential as per their request. Here are some of the key questions & answers from the interview -
Q. Hi, can you explain your current job profile & responsibilities in the organization?
As the Head of the Department, my job entails liaising with existing staff members, supervisors, and clients in order to achieve our goals. I also work on observing, analyzing, and offering suggestions to improve current operations, assisting the managers with recruitment & training processes and identifying key improvement areas & growth opportunities.
Q. What are the biggest challenges you face when it comes to talent management?
The biggest challenges I’d say is attracting and retaining top talent, developing employees’ skills and competencies, improving on their current performances and promoting a culture of learning and development amongst them.
Q. What are the most important metrics you use to measure the success of your talent management efforts?
Well there’s a fair few metrics I use and like to monitor like, employee engagement, retention & attrition rates, and performance metrics such as productivity and quality.
Q. What are the biggest pain points you experience when it comes to analytics?
Well I’d say data quality issues are very common and frustrating, the manner in which all the data is scattered makes looking for it very tedious and time consuming. There is still a lack of data integration across different systems, and most times it is really difficult to extract insights from available data.
Q. What are the most important metrics you use to measure the success of your analytics efforts?
I personally use ROI on the analytics investments, accuracy of available predictions and forecasts, and impact on business outcomes such as revenue growth or cost savings.
Q. What are the most important features you would like to see in a talent management & analytics tool?
The most important features I would like to see in a talent management & analytics tool include being ease of use so it takes minimal time to learn & onboard, ability to integrate with other systems & workflows such as HRIS or LMS, predictive analytics capabilities, and real-time reporting and dashboards.
Q. How do you currently manage your talent data?
Currently, I manage my talent data using Excel spreadsheets or Workday or SAP SuccessFactors, but yes mostly we just use Excel or Google Spreadsheets.
Q. How do you currently analyze your talent data?
Using Excel or other BI tools such as Tableau or Power BI.
Q. What are the biggest challenges you face when it comes to integrating data from different sources?
Like I said, data quality issues such as missing or incomplete data are the most common ones, lack of standardization across different systems, and difficulty in mapping data across different systems.
Those are all the questions I had, thank you for your time.
*end of interview*
This interview gave us great insights into the capabilities we could focus on for the product. After this, it was time to structure the application & think of potential ways to visualize the available datasets.
Making
Categorizing & Classifying Data
This project was a Data Visualization project as much as it was a User Experience Project for me. I worked with large amounts of corporate datasets that had to be presented simply & intuitively. That is why we started with categorizing data into different sections/pages according to their intent -
DIVERSITY & INCLUSIOM
Included Data related to cultural or gender characteristics like ethnicity, race, gender, etc.
WORKFORCE
Included Data specifically related to workforce attributes like employee age, experience, gender, etc.
TALENT ACQUISITION
Included Data related to recruitment & hiring candidates and the processes related to it like planning, sourcing, screening, interviewing, selecting, and onboarding talent.
ATTRITION
Included Data related to the trend of how many employees have left the organization in a specific time period and factors that may have caused them to leave.
PERFORMANCE
Included Data related to the individual and team performances of employees and overseeing that could be influencing employees to perform well or poorly.
COMPENSATION
Included Data related to the monetary compensation, revenue, financial benefits, costs, et cetera of the employees.
LEARNING & DEVELOPMENT
Included Data related to how many employees enrolled or didn’t enroll in the organization’s training program and to what degree it boosted the employees’ performances.
We also classified data into two types depending on the nature of it -
DESCRIPTIVE DATA
Data that summarizes and organizes the characteristics of a data set. It can describe the distribution and variability of the data using statistics, graphs, and tables.
PREDICTIVE DATA
Predictive data is the data that is used to make predictions about future outcomes or unknown events using historical data and statistical techniques.
Chart Selection
STACKED BAR CHART
To show data such as Male & Female Employees distribution or Category of Performers across a factor, etc.
TREEMAP
To show the absolute or percentage of a particular KPI across the Hierarchy of the Organization, i.e. Company, Country, Business Unit & Division.
PIE CHART
To show data such as Split between sources of applications or hires out of the total figure, HR-to-employee ratio in the organization, etc.
CANDLESTICK CHART
To show data including Pay Grades or Compensation Data across factors
TREND LINE CHART
To show time-centric data such as Attrition Rate, Performers' Trends, Internal Mobility Trends, etc.
WORD CLOUD
To show Textual, Keyword-based data such as data derived from Exit Surveys & Engagement Feedback Forms. Eg. of such a chart can be a 'Reasons for Leaving Word Cloud'.
HORIZONTAL BAR CHART
To show simple data such as Attrition vs Retention Rates, Male vs Female Hires, etc.
BAR & LINE CHART
To show the relationship between a trend & the absolute figures such as Active Headcount and Attrition Rate, No. of Employees Trained & Training Programs, etc.
SCATTER PLOT
To show data that is influenced by two variables like, Volume & Value Grid to see Employees' performances.
PYRAMID
To show different grades of employees and no. of employees in each Grade.
PIPELINE CHART
To understand recruitment flow starting from Requisition to Onboarding.
ORGANOGRAM
To understand the hierarchy of the organizational roles & departments and their relationships with Key KPIs such as Total Recruits, Attrition Rate, Vacancy Rate, etc.
Information Architecture
Here is the first level of Information Architecture (with New Additions) that we came up with. It was kept simple for the first review so we could get feedback on the more important & central aspects of the platform before getting into the details:
Wireframing
Wireframes for this platform were essentially how we would order the data charts & tables we wanted to showcase. But we also designed separate pages & flows for Uploading Files, Seeing Detailed Insights, et cetera.
Color Scheme & Typography
The Color Scheme was a combination of brand colors used by the Client and the primary colors used for the branding of our product (Talent PRO). Primary colors were different tones of Blue, used for the purpose of standing out on the interface. While we wanted to create a feeling of ease & spaciousness, we utilized white for backgrounds and shades of almost black (with varying opacities) for text. Buttons are Tabs were also made Blue for the purpose of 'popping out' from the rest of the UI.
A lot of secondary colors used were specific to the themes or demographics they represented. For example, Pink was utilized in the D&I section to create a clear distinction between Female & Male employees. Hence colors that already provoked a sense of familiarity were important to improve scalability & fast analysis.
Typography was kept to the Poppins family with varying weights to shift emphasis as per the need. The decision to not use too many fonts was taken to improve readability & speed up the development process.
1st Iteration & Feedback
Here are some screenshots from the First Cut of the prototype that was shared with the Client.
A lot of feedback surfaced as well which also included some sections we may have overlooked in the first version. Some of the key takeaways were -
A Summary View of Key KPIs & their Impact
Better application of the Predictive Model
Integration of Artificial Intelligence to drive Insights
A Quick View of the Data Health
A Separate Section for Performance for Sales Employees
2nd Iteration & New Additions (Another Lean Cycle)
After absorbing the feedback for another week & ideating on how to incorporate all of it into the design, we came up with the following new additions:
A Summary Page with an Organogram View and One or Two Important KPIs from different sections of the Platform drilled down to a Cross-level Department View
Predictive Insights to be generated & sent to the User as Notifications
A Dedicated Page for AI generated Nudges & Insights, categorized by Sections to drive decisions & action
A Data Health page to monitor the quality of the Data being fed to the model
A Performance For Sales page to view performance data for Sales Executives with different sets of KPIs & Terminologies.
Prototype
After adding the new elements & sections, the prototype was updated & shared with the Client who gave the go-ahead on the first Development cycle. Here is the updated version of the Prototype with screenshots of the new additions:
Conclusion & End Notes
Design is an iterative process. This remained true for our project as well. There were always new additions, corrections & processes that we could improve and add to make Talent PRO a more comprehensive, useful tool. But up until the second cut, we had a solid fundamental understanding of what we wanted to build for our Client and Users.
While the prototype lacked several key functionalities it was a great reference point to understand the User Flow & overall experience of the Platform.
Key Challenges for me in this project were to understand how people looked at data and how they wanted their data to look. Application of AI & Machine Learning was also particularly interesting in this project and the idea of using 'Nudges' to drive action was something new for me as a Designer.
In the present time, most sections of the platform are developed with real data being flown in. Design-wise, there haven't been any structural or functional changes, only new sections to incorporate new forms of datasets.
Thanks for reading!
Hope you enjoyed reading this User Experience Design Case Study.
To see more such projects, Click Here. To read more such articles, Click Here.