Analytics and Insights

Published on
Scene 1 (0s)

Analytics and Insights

Scene 2 (6s)

Realtime Dashboards

6003 26-Aug-19 16:38

6:13 ao PM a'2S2M9 Center RM validation E2E TAT



Driving business outcomes through continuous investments on analytical assets and transformation projects

Real time dashboards: Visualizations offer a combination of historic data and real-time information Provides insights on emerging trends and helps us manage the diverse data effectively. Enabling prompt decision making Dashboards are automatically updated and provide the user with instant access to critical data. Tracks the data that is important to user from various sources and presents that information in an easy-to-understand dashboard Interactive data visualizations: To analyze, track and report on the business data real time / near real time Governance Review (SLA + E-E metrics) through Advanced Power BI Dashboard All dashboards are developed Inhouse by our Advanced Power BI certified skilled resources, no added cost

Business Outcomes

Intuitive Visual Tool: Improved decision support through ( Proactive) meaningful insights

Improved Collaboration and Customer Satisfaction

Multiple Views Available Depending on User Role, Functions and Needs

Online Live Data: Sharing On The Go as a Continuous process with latest data

Circular flowchart

Presentation with bar chart


Alarm Ringing

Business Challenge

Scene 3 (55s)

Aug- Oct’18

Journey OF Customer experience IMPROVEMENT

Actions completed

Nov’18 -Jan’19

VOC in Mar’19 85%


Actions completed

Actions completed

VOC in Oct’18 68%


VBT 2.0 Updation Chat Audits Chat Refresher Training for P2P Upstream process Improvements to reduce volume Modern invoice implementation MPC tool enhancements Implementation of WHT tool vNext Chat and BOT

Actions Completed

Jul ’18

VOC in July 59%

AP QA Team & framework implemented Smart allocation /Prioritization of tickets via 1Q tool Re-validate PKT / FMEA/ DTP to cover all exceptions Smart comm & ‘what next’ checklist Skill Gap analysis and Training need identification Customer experience trainings initiated Customer Experience Awards instituted L1 & L2 teams formed

Restructure Process Knowledge tests to cover all exceptions Process Trainings based on Skill gaps Role Plays on floor WOW Customer Experience Awards launched and completed New Org Introduced ( FO BO Model ) Strengthened New Joiner Onboarding Process Comms Training batch commenced Pyramid re-levelling Continued reinforcement of customer service mindset Dedicated Chat Team formed Scenario based Job Aids created and teams trained

Process maturity model assessment Adopt leading practices and process fix. CX Training – 100% completed for PTP Front-office and Back-office. Continued reinforcement of customer service mindset Skit/Role Plays on floor continued FTR customer experience contest launched Digital Dash Boards for all towers Strengthened Hiring Process by implementing online REP Evaluation online test Design thinking workshop conducted for P2P to focus on Customer experience Leads Training – 1 Batch completed successfully, 2nd batch in progress SRT Search tool implemented Collaborative CRM workshop conducted - generated 88 ideas from Accenture

VOC in Jan’19 79%

VOC in Sep‘19 - 90 %



Current CSAT 90%

Scene 4 (2m 4s)

Intelligent T&E Solution

BUSINESS PROBLEM On an average,90000 T&E Claims processed per month. Significant audit effort is spent on auditing Time & Expense claims. which also leads to false positive claims Unknown non-compliant claims not caught by traditional rules Streamlining of travel booking and expense claims process

SOLUTION Predictive scoring engine solution which scores each claim and provide a highly predictive score for identifying noncompliance and waste & abuse Create an embedded predictive analytics solution (integrated with T&E tools) which can help clients in real/near real time prediction along with prescriptive actions

Predict non-compliance for every T&E claim to minimize leakage and significantly reduce audit effort



Covers 100% T&E claims <0.1 sec average time to score an expense 50%+ audit precision without compromising on quality (beat client expectation) 25%+ FTE reduction (better than client expectation)




Before T&E Intelligent (2017) January February March April May June July August September October November December 2553 2157 2685 2438 2867 4054 4803 3125 3184 4901 3855 4560 Post T&E Intelligent (2019) January February March April May June July August September October November December 992 1398 1748 1700 1785 1968 1949 2242 1849 1919 1968 2460



Scene 5 (2m 52s)

CRM Help Desk – Ticket Handling Invoice Payment Status Expense Payment Status Invoice Status PO Status update PO Cost Center Update PO Invoice Approver Change Close PO / Close Single line Change Cost Center Code Price decrease Quantity decrease 1600 QnA pairs across 20 categories 60+ Languages supported Bing Translate API ConnectMe technology (existing) Tickets Portal (existing) 1 2 3 4 5

Chat BOT




Self service Portal/Chat Bot

Live Chat

CRM Query

Current State


Email alias

Ticket creation

CRM Query

Before State

20% Query Reduction

BC CRM Volume FY19 P01 FY19 P02 FY19 P03 FY19 P04 FY19 P05 FY19 P06 FY19 P07 FY19 P08 FY19 P09 FY19 P10 FY19 P11 FY19 P12 FY20 P01 FY20 P02 FY20 P03 FY20 P04 FY20 P05 FY20 P06 FY20 P07 FY20 P08 FY20 P09 FY20 P10 5774 6169 5529 5298 4968 5205 4848 4841 5795 4785 4702 6314 5218 4588 4048 2923 2807 2934 2251 2609 3620 3164

Before implementation (Chat Bot) - 5205 Avg. vol

After implementation (Chat Bot) - 2901 Avg. vol



What’s Next: ChatBot for AP helpdesk

Scene 6 (3m 38s)


R based analytics engine for predicting performance Real time Power BI dashboard

Predictive model for PWT%: Weekly forecasting of PWT% is done with R based predictive analytics engine and Power BI for front end visualization. RAG status are added to gain immediate attention of the process manager. Integration of INPRO model: The INPRO model (currently in excel) would be integrated with Predictive dashboard, to give accurate view as to what group of invoices need immediate attention and what would be their impact on the PWT%, if not actioned.

:PWT Forecast Dashboard 'iscalYearDescr@.. 2020 Actuals MTD 97.02% ForetastMTD 9452% MPE 1.68% Target Period 2019 Fiscal Week Actuals(PWT) 9647% Forecast August 2019 FW 09

PWT predictive model


Database storing actual and predicted performance




% Payment on Time FY18 Q3 FY18 Q4 FY19 Q1 FY19 Q2 FY19 Q3 FY19 Q4 FY20 Q1 FY20 Q2 0.92 0.94 0.94 0.94 0.94 0.95 0.95 0.93 HQ FY18 Q3 FY18 Q4 FY19 Q1 FY19 Q2 FY19 Q3 FY19 Q4 FY20 Q1 FY20 Q2 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.88

Scene 7 (4m 26s)

Paid Within Term

Objective: To improve the BOI - Payment within term

Process as is


Improvement / tool

- Status

Manual email notification were sent by AP agents to the invoice approvers seeking approval Agents had to pull data manually with the invoices that are due for payment Manual follow ups must be done in a timely manner to ensure payment within terms

Eliminates manual data extraction and logic measurement through RPA, SSIS, SQL & .Net Automated emails being triggered at regular intervals for invoices that are due for payment Built Auto Bots to eliminate manual follow ups Smart email alerts and pings based triggered dynamically based on region’s business requirement

Timely email alerts to the approvers to follow up invoices that are due Smart Bot pings seeking approval of invoices Follow up done based on real time data

Significant improvement in TAT Improvement in payment within terms




Scene 8 (5m 6s)

Improvement in GLOBAL Discounts capture %

Automated Skype Notification Skype notification implemented for sending reminders to approvers for timely approvals 1 st reminder sent on 9th day from the invoice creation date 2 nd reminder sent on 10th day from the invoice creation date Ops team coordinate with tech team to ensure reminder notification sent on time Digitization Skype is integrated with SAP report – Parked Invoice Simplification Enhanced Reporting & Email alerts Robust Governance process implemented to track and ensure discount capture Regular report sent to approvers on pending items Automated process for sending reminder emails - 1st reminder (3rd day ), 2nd reminder (6 th day) and 3 rd reminder (10 th day) from creation date Escalation Matrix Implemented - Approver’s manager are copied 2nd and 3rd reminder 3 reminder Alerts sent Adhoc payment Run Review the invoices due for payment ,fall outs (due to exceptions) are identified and processed for payment Adhoc payment run executed on need basis to ensure on time payment On Need Basis Increase # of Vendor Analysis and inclusion of more vendors / invoices into Discount program All IT vendors added to the discount program Analysis and inclusion of more vendors

Opportunity Statement To avail the discount (2% to 5%), the invoice had to be approved and paid within discount payment terms. The client’s target for the new fiscal was to improve their discount capture rate to 97% percent to create additional savings

Increase in Global discount capture % from 87% to 99%

$750 Mn till FY20 Cum. Discount Captured

$5 Mn till FY19 Cum. Gain share


Replicating the Discount Capture for International as Phase 2

What’s Next?

% Discount Capture US FY18 Q3 FY18 Q4 FY19 Q1 FY19 Q2 FY19 Q3 FY19 Q4 FY20 Q1 FY20 Q2 0.96 0.98 0.97 0.97 0.98 0.99 0.99 1 HQ FY18 Q3 FY18 Q4 FY19 Q1 FY19 Q2 FY19 Q3 FY19 Q4 FY20 Q1 FY20 Q2 0.91 0.91 0.91 0.91 0.91 0.91 0.91 0.91

Scene 9 (6m 16s)

Capacity for next week is planned using the forecasting model Touch Per Hour (TPH) sub function wise made available Forecasted and capacity required is calculated for all sub functions

CRM server Forecasting Real time dashboard Design

Phase 2 (In Progress) Developing an optimal staffing patterns (Shift / hourly flows / capacity required)

CRM Forecasting and Capacity model

Month August v OFTER.equired 'Forecast 12 10 August 19 Touched of Tickets Sub Name - Actual vs Touched Tickets - Actual vs Forecast Sub Function QM -Europe v Week Week>3 Aug.„ v August 20 Date 4/1/2019 9/30/2019 Actual Volume 3,492 Forecast Volume 3,157 128 August 22 Forecast FTE 12.14 581 August 23 TPH 6.50 Forecasted Volume and FTE Vs Actual Volume August 21 700 500 100

Scene 10 (6m 44s)


A close up of a logo Description generated with very high confidence



Ticket Received in 1Q tool

Sorting and allocation (based on rules)

Processing ticket based on prioritization model


Reporting & Power BI Dashboard

1Q – Customized UI & PBI

1Q – Customized UI & PBI

Team leads extract the report from CRM, categorize tickets manually based on different category and allocate to agents on daily basis Effort required for sorting and allocating consumes more time Building a predictive model to prioritize tickets based on SLA’s and identify high risk tickets (VIP) for “Reducing the escalation” by prioritizing and allocating tickets This led to: Improving “ Voice of customer ” Reduce escalation from client and supplier Improving the tickets Re-open% by providing end to end resolution Focusing on FTR to reduce the effort spent on multiple touches

1Q workflow tool - Managing Ticket Allocation and processing