HRMS Chatbot - Case Study
Talal Abu-Ghazaleh Global (TAG.Global) is the leading group which has a presence in several countries and is into the business of providing professional services across various domains of Accounting, External Audit, Internal Audit, Corporate Governance, Taxation, Educational Consultancy, Economic and Strategic Studies, Management Advisory Services, Professional and Technical Training, Technology Transfer, Project Management, Investors to name a few.
Website: www.tagorg.com
Brand: Talal Abu-Ghazaleh
Industry: Professional service and education
Location: Jordan, Oman
Core Platform: Web
Programming Language: ReactJS, NodeJS,
Framework: Botpress, NLP, pyAudio
Employees
Worldwide
Challenges
- Handling the multiple systems to integrate into the solution
- Authenticating employee data
- Maintaining the privacy and security of the system, data, etc.
- The implementation of the Arabic language was a bit difficult.
- BOT Press, cheap solution but less documented.
- A user interface to compile with Bot Press support
- Cost-effective robust Voice BOT integration
Strategic Approach
The client's requirements were feasible, the only challenge was to implement audio interaction instead of text input with cost-effectiveness. Also, We identified that BOT Press was supporting Multiple BOT session, but the client's UI requirement was against the standard of BOT Press UI guidelines. so, we Integrated the Voice Functionality with cost-effectiveness
The steps we took to understand the client business, its target audience, the problem, and challenges the client is facing, the goals they need to achieve were:
Before the actual project started we have gathered detailed requirement from a client about how their existing process of HR and Employee interaction, which process they want to migrate to BOT, business rules, Work Flow (Sequence of Questions and Answer), etc, to ensure that we were building the right application Analyzing the different software they were using and identifying the areas where they were facing issues to answer the repetitive questions on a day to day basis.
Before the actual project started we collected the following documentation to ensure we are building the right application:
- RFP
- SOW
- Design documents
- Sample Data/Physical Printed Forms of current manual process
- Organization hierarchy and its possible accessibility
- Solve the day to day activities of HR and resolve employee's questions/concerns efficiently
- Drive efficiency in its reading system
- Collecting the resume of the prospective employee through a chatbot
- Automating HR functions and processes with the help of chatbot and scheduling interviews
- Automate and reduce common activities of HR through BOT
- Collect the resumes for recruitment and interview schedules.
- Getting done the activities related to FAQs, HR Policies, Payout/Salary queries, leave request/approval
- Support Voice-based interaction from Operator to BOT.
- Reduce HR routine work along with maximum HR process automation
Project Development
The project team consisted of developers
BOT Press developer
React - NodeJS developers
Backend Developers
API Developer
Quality Analyst
Project Manager
Designer
Business Analyst
Scope:
- Leave policies
- Salary
- Pending leaves log
- Questions on HR policies
- Leave request
- Status on the leave request
- Apply for Job
- Interview Scheduling
Timeline: 5 Months
Project Highlights
Application Features
- Voice and Text Chat BOT
- QnA
- Leave Policies
- HR Policies
- Leave Approval
- Payroll
- Job Posting
- Apply for Job
- Interview Scheduling
Key Highlights
- Browser-based
- Use of Open-source libraries only
- 3rd party functionalities should be cost-effective
- Accuracy level 70% or higher at an initial level
Key Takeaways and Learnings
BOT Press is a good open-source BOT development framework.
Custom UI and Multiple BOT is complicated with the BOT Press framework
Voice command support is costly in terms of time/money for all browser compatibility
Reduction cost
Increase in efficiency
Business Impact
- By developing this application, the company was able to see a 30% increase in efficiency for repeated questions
- 15% reduction in cost due to automated recruitment and interview process.