QMS Software Development
Budget: ₹150,000 – ₹250,000 INR
Execution Plan:
Sampling
Sampling will be conducted in two levels:
Creation of Sampling Pool:
Create a sampling pool (B) from Ops production (A) daily at the process level (e.g., AR, CO, PO) based on predefined, dynamic sampling logic as per business requirements.
Examples for AR process:
High aging claims (<30 days aging excluded).
High-value claims (criteria defined by the client).
Complex/Denied claim categories (exclude in-process, paid, zero-out claims; include "paid but not posted" claims aged 90+ days).
Sub-process categories (e.g., Exception/Tasking).
Insurance and time-stamped claims (if available).
Duration-based sampling (time taken to complete the claim).
Examples for Posting process:
Manual batches.
Denial/Correction batches/entries.
ERA batches sampled minimally or excluded as per requirements.
Sampling interface for QA Manager to:
Set weightages (e.g., 20%, 30%).
Enable multiple selection options for modifications.
Note: Ensure no duplicate data in the sampling pool.
Sample Picking/Throwing:
Sequentially pick samples from sampling pool (B) and assign to QAs, masking agent details.
Define sampling percentages based on agent stack ranking:
Stack rank based on monthly quality scores.
Sample percentage examples (dynamic):
Quality Score >=98%: 5%
Quality Score >=95% - <98%: 10%
Quality Score >=90% - <95%: 15%
Quality Score <90%: 20%
Adjust sampling% dynamically by QA Manager as needed.
Address insufficient data by sampling randomly from Ops inventory (A), excluding duplicates.
Unused samples retained for future audits (buffer).
Sequential sampling ensures diverse agent and queue representation unless "focused audit" is specified.
Additional Sampling Features:
Focused Audit:
Agent-wise sampling based on tenure.
Queue-wise sampling for specific priorities.
Same-Day Sampling:
Provision for real-time sampling for coding/CP audits.
Non-IWS Clients:
Option to upload Ops inventory in a template for sampling.
Reports:
Maintain exportable reports for Ops inventory (A) and Sample Pool (B).
Exportable sampling plans (client/process-wise) or read-only access for Ops teams.
Audit
QAs will receive samples sequentially, masking agent details.
QAs can:
Select client/sub-process for audit.
Audit within a 5-day window.
Capture audit details using pre-defined templates.
Audit types include:
Data/Call audit.
Error category, sub-category, error type, impact, RCA.
Allow QA modifications on the same day. Managers can update audit results with justification.
Time Stamping:
Track time spent per account for capacity utilization analysis.
QA Productivity Monitoring:
Enter QA targets/day to measure productivity, utilization, and incentive plans.
Audit Tool Uploads:
Facilitate urgent Ops requests with upload capabilities for external audit spreadsheets.
Additional Features:
Real-time QA productivity dashboard (parked for future).
Error Marking & Communication
Trigger error notifications to respective agents, masking PHI, and copying Ops/Quality supervisors.
Process Gap Errors:
Mark errors due to process issues under a "Process Gap" dummy agent to maintain quality score consistency.
Require agents to correct errors with acknowledgment.
QA revisits errors within 48-72 hours to verify corrections. Escalate additional errors if uncorrected.
Error Challenges:
Agents can challenge errors within 48 hours with notes.
QA team can reject/reverse challenges within 48 hours, closing disputes within 96 hours.
Maintain error challenge logs for audit trails.
Escalate unresolved disputes to discussions outside the tool.
Rebuttal
Implement a transparent rebuttal process for disputes:
Flow: Agent → Ops TL → QA Team.
Timestamps for all actions (error marking, challenges, rebuttals).
Resolve disputes within defined SLAs (e.g., 96 hours).
Reporting
Generate exportable reports for:
Sampling plans.
Quality scores.
Error logs and resolutions.
QA productivity metrics.
Re-Audit Module
Retain samples for re-audit.
Track historical audit changes and follow-ups.
Escalate unresolved or critical errors for second-level review.
System Intelligence
Error Pattern Detection:
Automatically park claims with similar errors for review.
Convert standardized error comments into machine-readable formats for pattern identification.
Backend Automation:
Identify repetitive errors across claims for proactive interventions.
Development Requirement:
UI/UX Design
Backend: Node.js (Express)
Database: MongoDB
Frontend: React.js
Messaging and Background Task: RabbitMQ
Sampling
Sampling will be conducted in two levels:
Creation of Sampling Pool:
Create a sampling pool (B) from Ops production (A) daily at the process level (e.g., AR, CO, PO) based on predefined, dynamic sampling logic as per business requirements.
Examples for AR process:
High aging claims (<30 days aging excluded).
High-value claims (criteria defined by the client).
Complex/Denied claim categories (exclude in-process, paid, zero-out claims; include "paid but not posted" claims aged 90+ days).
Sub-process categories (e.g., Exception/Tasking).
Insurance and time-stamped claims (if available).
Duration-based sampling (time taken to complete the claim).
Examples for Posting process:
Manual batches.
Denial/Correction batches/entries.
ERA batches sampled minimally or excluded as per requirements.
Sampling interface for QA Manager to:
Set weightages (e.g., 20%, 30%).
Enable multiple selection options for modifications.
Note: Ensure no duplicate data in the sampling pool.
Sample Picking/Throwing:
Sequentially pick samples from sampling pool (B) and assign to QAs, masking agent details.
Define sampling percentages based on agent stack ranking:
Stack rank based on monthly quality scores.
Sample percentage examples (dynamic):
Quality Score >=98%: 5%
Quality Score >=95% - <98%: 10%
Quality Score >=90% - <95%: 15%
Quality Score <90%: 20%
Adjust sampling% dynamically by QA Manager as needed.
Address insufficient data by sampling randomly from Ops inventory (A), excluding duplicates.
Unused samples retained for future audits (buffer).
Sequential sampling ensures diverse agent and queue representation unless "focused audit" is specified.
Additional Sampling Features:
Focused Audit:
Agent-wise sampling based on tenure.
Queue-wise sampling for specific priorities.
Same-Day Sampling:
Provision for real-time sampling for coding/CP audits.
Non-IWS Clients:
Option to upload Ops inventory in a template for sampling.
Reports:
Maintain exportable reports for Ops inventory (A) and Sample Pool (B).
Exportable sampling plans (client/process-wise) or read-only access for Ops teams.
Audit
QAs will receive samples sequentially, masking agent details.
QAs can:
Select client/sub-process for audit.
Audit within a 5-day window.
Capture audit details using pre-defined templates.
Audit types include:
Data/Call audit.
Error category, sub-category, error type, impact, RCA.
Allow QA modifications on the same day. Managers can update audit results with justification.
Time Stamping:
Track time spent per account for capacity utilization analysis.
QA Productivity Monitoring:
Enter QA targets/day to measure productivity, utilization, and incentive plans.
Audit Tool Uploads:
Facilitate urgent Ops requests with upload capabilities for external audit spreadsheets.
Additional Features:
Real-time QA productivity dashboard (parked for future).
Error Marking & Communication
Trigger error notifications to respective agents, masking PHI, and copying Ops/Quality supervisors.
Process Gap Errors:
Mark errors due to process issues under a "Process Gap" dummy agent to maintain quality score consistency.
Require agents to correct errors with acknowledgment.
QA revisits errors within 48-72 hours to verify corrections. Escalate additional errors if uncorrected.
Error Challenges:
Agents can challenge errors within 48 hours with notes.
QA team can reject/reverse challenges within 48 hours, closing disputes within 96 hours.
Maintain error challenge logs for audit trails.
Escalate unresolved disputes to discussions outside the tool.
Rebuttal
Implement a transparent rebuttal process for disputes:
Flow: Agent → Ops TL → QA Team.
Timestamps for all actions (error marking, challenges, rebuttals).
Resolve disputes within defined SLAs (e.g., 96 hours).
Reporting
Generate exportable reports for:
Sampling plans.
Quality scores.
Error logs and resolutions.
QA productivity metrics.
Re-Audit Module
Retain samples for re-audit.
Track historical audit changes and follow-ups.
Escalate unresolved or critical errors for second-level review.
System Intelligence
Error Pattern Detection:
Automatically park claims with similar errors for review.
Convert standardized error comments into machine-readable formats for pattern identification.
Backend Automation:
Identify repetitive errors across claims for proactive interventions.
Development Requirement:
UI/UX Design
Backend: Node.js (Express)
Database: MongoDB
Frontend: React.js
Messaging and Background Task: RabbitMQ