1 20170616 Automation in Financial Services_ADETEM.pptx Automation in Financial Services Laurent Doucet Club Banque, Finance Assurance Mardi 20 juin 2017 Ère de l’intelligence artificielle… et si la machine nous remplaçait ? En partenariat avec :
1 20170616 Automation in Financial Services_ADETEM.pptx
Automation in Financial Services
Laurent Doucet
Club Banque, Finance Assurance
Mardi 20 juin 2017
Ère de l’intelligence artificielle… et si la machine nous remplaçait ?
En partenariat avec :
2 20170616 Automation in Financial Services_ADETEM.pptx
Contents Page
This document shall be treated as confidential. It has been compiled for the exclusive, internal use by our client and is not complete without the underlying detail analyses and the oral presentation.
It may not be passed on and/or may not be made available to third parties without prior written consent from .
© Roland Berger
A. Overview 3
B. Use cases 11
C. Detailed case studies 22
D. How to move forward? 33
3 20170616 Automation in Financial Services_ADETEM.pptx
A. Overview
4 20170616 Automation in Financial Services_ADETEM.pptx
Software-driven automation has the potential to raise efficiency to the next level across industries
Major efficiency levers over time
Source: Press research; Roland Berger
Today artificial intelligence is where the internet was in 1996"
We are entering a new phase in world history – One in which fewer and fewer workers will be needed to produce the goods and services for the global population"
Business process automation
2015 2005 1995
Automation of business processes using software robots
Quellen:
"Today AI is …"
http://www.merantix.com/
http://raceagainstthemachine.com/
By Erik Brynjolfsson
(http://twitter.com/erikbryn) and Andrew
McAfee (http://twitter.com/amcafee)
From off-shore to no-shore
(http://www.opuscapita.com/blog/2015/medium-sized-companies-don%E2%80%99t-outsource-they-do-and-the-trend-goes-upward)
From offshore to no-shore"
Business process outsourcing/offshoring
Outsourcing of operations and responsibilities to service providers in countries with lower labor costs
Business process re-engineering/management
Analysis and design of business processes within an organization to reduce non-value-adding work
Overview A
5 20170616 Automation in Financial Services_ADETEM.pptx
Software-driven automation techniques
Automation platforms offer higher flexibility than traditional scripting – Emerging machine learning opens up a whole new world
A Overview
Source: Roland Berger
Automation technique
Tailored software and scripting
0 Robotic process automation/ Automation platforms
1 Artificial intelligence (machine learning)
2
Data characteristics
Structured (rigid) Structured or patterned Unstructured and unpatterned large data sets
Examples > Complex reports in SAP
> Tailored workflow tools
> Automation of IT operations/tickets
> Aggregation of data from multiple systems
> Recognition of security threats from deviation of normal behavior
> Self-driving cars learning from observing humans
Description > Scripts or tailored (enterprise) software to support a specific process or workflow
> Rigid processes and high programming/testing effort are typically required
> Tools and platforms that help to automate and orchestrate repetitive processes across existing systems
> Software interfaces or non-invasive approaches mimicking human behavior
> Advanced algorithms that can handle ambiguity – Self-learning replaces need for prescriptive rules
> Systems that adapt their behavior based on observing humans
Traditional approach Emerging technology, vast potential
Tools available, usage increasing
Flexibility Low Medium High
Implementation effort
High Low Medium
1/2 Deep dives presented in this document
6 20170616 Automation in Financial Services_ADETEM.pptx
RPA and automation platforms promise to automate repetitive tasks easily without the costs and limited flexibility of tailored software
Robotic Process Automation & Automation platforms
1) What You See Is What You Get
Details Use cases
Robotic process automation
> Moving files and folders
> Scrapping data from the web
> Extracting and reformatting data into reports and dashboards
> Merging data from multiple places
> Software that mimics human behavior at a computer, e.g., non-invasive software to automate repetitive tasks
> RPA either aims to replace human labor or assist a human worker to improve efficiency
> Very easy to set up and adjust making deployment feasible even for one-off tasks, e.g., WYSIWYG1) interfaces
> Dedicated systems that aim to make automation easy
> Scripting/orchestration across applications using one common platform/language
> Connected to existing (legacy) systems via software interfaces and APIs
> Continuous improvement approach with constant creation and adaption of scripts
> Monitoring, escalation, and analysis to support operations
Automation platforms
> Update meta data in cloud environments
> Deployment of virtual machines
> Auditing and reporting the health of IT stacks in real time
> Smart City systems (e.g., smart parting systems)
> Allocation of cores and RAM for simulations on supercomputers
Source: WorkFusion; Roland Berger
1 A
RPA / Automation platforms – Description and examples
7 20170616 Automation in Financial Services_ADETEM.pptx
RPA is expected to be one of the next major disruptions to come with sharp impact on existing way of doing business
Automation is threatening to replace swats of white-collar workers, much as mechanical robots have displaced blue-collar workers on assembly lines.
Wall street Journal
Robotic process automation will be the next big
disruptor.(…) Every organization will find the
combination that is right for it. But getting ahead of
this curve is paramount because RPA is here to stay.
Tanvir Khan, Dell
Anything that could give rise to smarter-than-human
intelligence — in the form of Artificial Intelligence,
brain-computer interfaces, or neuroscience-based
human intelligence enhancement — wins hands down
beyond contest as doing the most to change the
world. Nothing else is even in the same league."
Eliezer Yudkowsky, Co-Founder and Research
Fellow, Machine Intelligence Research Institute
Computer coded software that:
Walking, talking, independent robots, replacing humans in all their capabilities
What
it is What
it is not
Physical machine processing physical things
A software with artificial intelligence or voice recognition with reply functions
Definition Software that simulates a 'virtual person' and interacts with existing application software through rule-based tasks in the same way humans would do
Replace humans in performing repetitive rule-based tasks, use logic to model decisions in the process
Interact with any application or system and can work on multiple systems
Process transactions, manipulate data, triggers responses and communicate with other digital systems
✓ ✗
RPA in a nutshell
Robotic Process Automation & Automation platforms 1 A
Source: Roland Berger
8 20170616 Automation in Financial Services_ADETEM.pptx
AI relies on machine learning techniques for problem solving – A breakthrough in deep learning enabled AI for practical applications
Search and optimization
Constrain satisfaction
Local reasoning
Control theory
Probabilistic reasoning
Machine learning
Reinforcement learning Random forest
Support vector machines
Bayesean networks
Genetic algorithms
Deep Learning
Association rule learning
Decision trees
Problem solving techniques
Artificial intelligence – Terminology classification
Source: IBM; Roland Berger
Artificial intelligence
> Artificial intelligence extends cognitive computing by not only suggesting solutions to problems, but also by making actual decisions based on the results of data analyses
> Example: Self-driving car that analyses the environment and decides to break, accelerate, or change lanes
Cognitive Computing
> Cognitive computing supports people in making decisions by analyzing large amounts of (unstructured) data and suggesting solutions to problems
> Cognitive computing systems only support the decision making, the actual decision is taken by humans
> Example: System that analyses patient data and suggests potential treatment options including advantages and disadvantages to doctors
Artificial intelligence (machine learning) 2 A
Most important problem solving technique for AI
9 20170616 Automation in Financial Services_ADETEM.pptx
Artificial intelligence – Simplified history
It took close to 60 years and many so-called "winters" of stagnation for AI to reach today's state
Source: Press research; Roland Berger
1956-1974
> Solving problems in a specific domain of knowledge by using rules derived from experts
> Potential use cases were:
– Identification of chemical compounds from spectrometer data
– Diagnosis of infectious blood diseases
> Understanding and processing natural language, e.g., translating from English to Russian
Natural language
Reasoning as search
> Finding a problem solution by searching for the answer
> Focus on artificially simple situation, e.g., block worlds consisting of colored blocks of various shapes and sizes
Microworlds
> No translation of words with context-dependent meaning possible
> No extrapolation outside micro- worlds
> Complexity exceeding avail. computing power
Issues leading to disappointment of ambitions and stop of funding
AI winter
Expert systems
> No synergies in creating expert systems for different domains – Proprietary algorithms for each system required
1980-1987 1980-2012
Intelligent agents
> Isolation of problems and finding verifiable and useful solutions
> Common language allowing interaction with economics and control theory
New advanced tools
> Utilization of new tools like
– Bayesean networks
– Stochastic modeling
– Neural networks
– Evolutionary algorithms
Break- through 2012
Deep learning
> Utilization of multi-level neural networks to solve complex problems like picture and speech recognition
Issues leading to disappointment of ambitions and stop of funding
AI winter
Artificial intelligence (machine learning) 2 A
10 20170616 Automation in Financial Services_ADETEM.pptx
Artificial intelligence – Projects of global technology companies
As a response to AI's success in recent years, most global technology companies have made it one of their key priorities
Source: Press research; Company information; Roland Berger
Google IBM Facebook
> The Google Brain project investigates deep learning since 2011 and developes TensorFlow, an open source software library for artificial intelligence
> In 2014, Google acquired DeepMind Technologies, the company that later developed the AlphaGo program
> IBM started to develop its cognitive computing system Watson in 2005 and added deep learning algorithms after its commercialization in 2014
> Based on Watson, IBM offers solutions for R&D projects in the pharma, publishing, and biotechnology industry, self-service applications, as well as enterprise analytics
> FAIR was founded in 2013 and developed several deep learning algorithms used for photo tagging and text translation as well as extensions to Torch, an open-source library for AI development
> In 2015, Facebook acquired Wit.ai that currently develops its personal assistant "M"
"Our deep learning tool has now been deployed in many environments, particularly across Google in many of our production systems"
"Watson is the the biggest, most important thing I’ve seen in my career and is IBM’s fastest growing new business in terms of revenues"
"We’re trying to build more than 1.5 billion AI agents – One for every person who uses Facebook or any of its products"
Artificial intelligence (machine learning) 2 A
11 20170616 Automation in Financial Services_ADETEM.pptx
B. Use cases
12 20170616 Automation in Financial Services_ADETEM.pptx
Application of RPA and augmented intelligence by activity type and impacts
Source: Roland Berger
Robotic Process Automation and Augmented Intelligence apply to different perimeters
Easily robotized
activity
Simple
and rule-
based
Complex
and
judgment
based
Unstructured Structured
Sim
plic
ity
Structure
> Software that simulates a 'virtual person' and interacts with existing application software through rule-based tasks in the same way humans would do
Robotic Process Automation
Focus on
maximizing
robotization and
interfacing with
human
interaction
Focus on
structuring
processes to
enable
robotization
Complex activity
needing human-
like decisions
-
Augmented
Intelligence
sweetspot
> Partially automates operations and enhances complex decision making through solutions combining Natural language processing, machine learning and hypothesis generation
Artificial Intelligence
Quality
Speed
Cost
1
2
3
> Reliability: 100% accuracy > Improved customer satisfaction > Better decision making thanks to
increased focus of staff on more added value tasks
> Solutions working 24/7 > Enhanced processing speed > Increased capacity to handle
volume in back office leading to less demand failure in front office
> Quick payback ( Typically 12-24 months)
> Reduced labor costs > Limited investment needed given
smart interfacing with existing IT infrastructure
1
Use cases B
13 20170616 Automation in Financial Services_ADETEM.pptx
In most industries, maximum AI and RPA potential is reached on back / middle-office activities and selected support functions
AI RPA Human
Source: Roland Berger
Overview of potential use of RPA and AI technologies by function
Front-office Back-office Middle-office Support functions
Au
tom
atio
n p
ote
nti
al
> Highest automation potential in Back
and Middle office activities driven
by:
– input data already digitized to a
large extent
– industrialized processes with
clear rules
– proliferation of IT systems and
tools (CRM, sales, claims, etc.)
> Several support functions involving
data processing can be automated
to a large extent (e.g. accounting,
controlling, payroll, etc.) – unlike
functions involving more creativity
and human interactions (marketing,
communication, recruitment,…)
Use cases B
14 20170616 Automation in Financial Services_ADETEM.pptx
Overview of vendors (selection)
A growing number of vendors is offering software solutions for business process automation
Source: Company information; Roland Berger
Typical use cases
Solution providers
> Automated data transfer between different systems, e.g., between CRM and ERP systems.
> Error detection for large data arrays like transaction matching and account reconciliation
> Automated cybersecurity incident response
> Self-learning of ability to distinguish between different types of documents, e.g., between invoices, claims and questions
> Chat and voice bots with ability to process natural language and to answer automatically including clarification questions if necessary
Use cases B
Robotic process automation/ Automation platforms
1 Artificial intelligence (machine learning)
2 Automation technique
15 20170616 Automation in Financial Services_ADETEM.pptx
The number of growing use cases for RPA, automation platforms, and AI confirms the large potential of software-driven automation
Overview of selected use cases
Source: Press research; Company information; Roland Berger
Banking process automation (RPA)
> Automation of review processes for banking transactions
Use cases B
Software service automation (RPA)
> Automated handling of software requests
Robotic process automation/ Automation platforms
1 Artificial intelligence (machine learning)
2
Automated cyber attack response (RPA)
> Automated reaction to cyber threats
IT incident handling (Automation platform)
> Automated solving of standard IT incidents, esp. L0 and L1
> 85% automation of the SSI1) process in global banking
Call center automation
Automated email processing
> Automated categori-zation of emails incl. recognition of key data
Automated claims processing
> Automated import of data from claims in database
Automation of SSI process in banking
Recruiting process automation
> Automated candidate search and prequalification
> Automated handling of incoming customer calls
Automated energy management
> Automated optimi-zation of data center power consumption
1) Standard settlement instruction
16 20170616 Automation in Financial Services_ADETEM.pptx
Automation of banking transactions / IT services / reaction to cyber threats
RPA can be used to automate standard banking tasks, IT services, and the handling of cyber threats
Source:
http://www.blueprism.com/case-studies
Source:
http://thoughtonomy.com/computacenter-deploy-virtual-workers-in-service-desk-2/
Source:
http://ayehu.com/cyber-security-incident-response-automation/eyeshare-for-automated-cyber-security-incident-response/
Robotic process automation & Automation platforms 1 B
Problem
> Each day more than 2500 high-risk bank accounts with insufficient funds have to be reviewed manually
✓
Approach
> RPA is used to automate the process based on predefined rules
> Software accesses the bank's core systems and does not require any system changes
Advantage
> 80% reduction of processing costs
> Process time reduction by more than 50%
> Increase of consistency
Problem
> High workload for service desk staff due to manual procedures shifts focus away from their individual customer service tasks
✓
Approach
> 95% of key user administration tasks are offered via an self-service portal by utilizing RPA
> After service request, RPA performs task by emulating a human user
Advantage
> Self-service portal enables 24/7 execution of key activities
> Desk staff can focus on customer service instead of manual intensive tasks
Problem
> IT personnel can only hardly handle increasing volume and high speed of cyber attacks
✓
Approach
> RPA is used for
– automated log-out and password reset in case of multiple simultaneous logins
– isolation of client from LAN in case of malware detection
Advantage
> Decreased response time to cyber incidents
> Reduced workload for IT personnel
Source: Press research; Company information; Roland Berger
17 20170616 Automation in Financial Services_ADETEM.pptx
Artificial intelligence (machine learning) 2 B
Call center automation
Amelia understands, learns, and adapts to natural language to handle service desk and expert advisory tasks
Cognitive agent Amelia
Understanding information
> Amelia understands written and spoken language including contextual information
> She is able to understand the user's mood
Learning
> Amelia learns from live interactions
> If she cannot solve a problem, she hands it over to an employee and learns by listening to him
Service desk / call center support
✓ Problem
> The IT service desk of a large media company needs to handle more than 65,000 calls per month which leads to high workload
Approach
> Amelia learned to take 64% of the incoming calls through observational learning
Advantage
> Reduction of staffing requirements from 76 to 32 FTEs
> Reduction of the mean time to resolve an issue from 18.2 to 4.5 minutes
> Reduction of the average speed of an answer1) from 55 to 21 sec.
✓ Problem
> Equipment troubleshooting requires large amounts of knowledge
Approach
> Amelia learned from machine manuals and company policies and provides guidance to engineers
Advantage
> Improved equipment trouble-shooting in complicated situations
Expert advisor for field engineers
1) Average time it takes for a call to be answered, includes time in waiting loop and duration of time in which the agents phone is ringing
Source:
http://leoforce.com/product.php
Source: Company information; Roland Berger
18 20170616 Automation in Financial Services_ADETEM.pptx
AI solutions have already started to be implemented in Financial Services to optimize Middle Office activities (1/2)
Underwriting Contracts Management – Middle Office
Contracts Management - Operations
Claims Management
Solution prov. Year
Use cases in Financial Services – Middle Office
2016
2015
2016
2016
2015
2016
Service
Assistant virtuel : Réponse aux questions des chargés de clientèle dans le domaine de
l'assurance
Moteur de recherche intelligent multi-source (textes, images, média sociaux,…) : analyse des données client et identification des moments de vie
Moteur de recherche intelligent : Recherche de toutes les données disponibles (structurées et non structurées) pour construire une vision client synthétique, 360° en temps réel
Agent conversationnel : Réponse en direct aux questions des clients (ou transfert vers un gestionnaire si la complexité est trop élevée)
Moteur de recherche multi source comprenant le langage naturel : anticipation de la volumétrie des motifs d'appels au support client, pour jour la FAQ / page d'accueil en anticipation
Analyseur d'email clients: Détection de l'intention et prise automatique des rendez-vous commerciaux / réponses à certaines demandes (ex : transmission d'attestation d'assurance)
Agent conversationnel intelligent répondant aux questions des clients
Country
2014
Client
RBS
Source: Analyses Roland Berger
Artificial intelligence (machine learning) 2 B
19 20170616 Automation in Financial Services_ADETEM.pptx
AI solutions have already started to be implemented in Financial Services to optimize Middle Office activities (2/2)
Use cases in Financial Services – Middle Office
2015 Hitachi (TBC) Moteur de recherche support aux gestionnaires (call center auto) : Recherche de réponses aux questions clients – grâce à la compréhension de la voix. Gain de 15% sur les temps de communication
2013 Assistant virtuel: réponse aux questions des clients sur les produits et services
Non communiqué 2015 Moteur de recherche intelligent : Automatisation du processus de KYC (recherche, agrégation, et vérification des données clients )
2012 Assistant vocal KAI répondant oralement aux demandes des clients concernant leur compte bancaire
2016 Agent conversationnel : Réponse aux questions des clients (après avoir été entrainée en interne au sein du "helpdesk" du service informatique)
Assistant virtuel : Réponses aux questions des clients, intégré à l'application mobile de la banque
2016
Source: Analyses Roland Berger
Artificial intelligence (machine learning) 2 B
Underwriting Contracts Management – Middle Office
Contracts Management - Operations
Claims Management
Solution prov. Year Service Country Client
20 20170616 Automation in Financial Services_ADETEM.pptx
AI solutions have already started to be implemented in Financial Services to optimize Operations activities
Use cases in Financial Services – Operations
Analyseur d'emails : Identification du contenu des emails client (intention, urgence) et lancement d'actions pour certaines intentions (ex : pré remplissage des champs de virement)
2016
2014 Analyseur d'emails : Identification du contenu des emails client, routage vers le service compétent, proposition de réponses automatiques, aide à la réponse,
Agent virtuel conversationnel. Réponse aux questions des clients et réalisation d'opérations : virement, analyse de dépenses,… via une interface de "chat" intégré à l'application
2016
Source: Analyses Roland Berger
Artificial intelligence (machine learning) 2 B
Underwriting Contracts Management – Middle Office
Claims Management
Solution prov. Year Service Country Client
Contracts Management – Operations
21 20170616 Automation in Financial Services_ADETEM.pptx
AI solutions have already started to be implemented in Financial Services to optimize Claims Management activities
Use cases in Financial Services – Claims Management
Aide à la détection de fraude : identification de profils de fraudeurs basé sur l'analyse des données clients
Confidentiel
Aide à la détection de fraude : construction en temps réel d'un score qualifiant le caractère suspect ou non des déclarations de sinistres envoyées et de qualifier les types de fraudes potentiel
Confidentiel
Digitalisation du traitement des sinistres, grâce à la reconnaissance du langage écrit, l'auto remplissage de formulaires, et l'interface avec de nombreux systèmes de gestion
Global
Application mobile de déclaration des sinistres : Scan des pièces justificatives, détection automatique des fraudes
Traitement automatisé des sinistres via la reconnaissance du langage écrit, l'extraction des informations, la comparaison des informations vs. termes de l'assurance
COGITO
Underwriting Contracts Management – Middle Office
Contracts Management - Operations
Claims Management
Global 2015+
2016
2016
2015+
2015+
Source: Analyses Roland Berger
Artificial intelligence (machine learning) 2 B
Solution prov. Year Service Country Client
22 20170616 Automation in Financial Services_ADETEM.pptx
C Detailed case studies
23 20170616 Automation in Financial Services_ADETEM.pptx
RPA was the preferred option in a cost reduction exercise for the finance function of a global insurance company
Context Approach
> Leading insurance company
with operations in +40
countries and ambitious
growth and profitability targets
> Finance team of ~200 FTEs
with Accounting team
representing 60% of finance
staff. Limited use of SSC or
offshoring so far
> Recent merger in the group
has set expectations for
synergies in the Finance
function
> Overall cost reduction of 20% set for
the finance function, combined with a
necessity to reduce headcount
> Target cost reduction of 30+% for
accounting as it was identified as the
area with most potential
> Integrate finance organizations of
recently merged companies
> Detailing of activities within
accounting (80+ activities for 125
FTE)
> Definition of baseline volume of
workload (#FTE) per activity
> Evaluation of potential for
process automation and
offshoring for each individual
activity
> Consolidation and challenge of
results from a general perspective
to increase the level of
offshoring/automation while
maintaining local oversight
Source: Roland Berger
Case study – Finance RPA in Insurance
RPA case study – Finance function automation C 1
24 20170616 Automation in Financial Services_ADETEM.pptx
Criteria of savings potential & digital feasibility drive the decision for automation together with local constraints
Source: Roland Berger project experience
Should the activity be robotized or off-shored?
Possible gains from robotization?
• Is the process (relatively) stable over time with frequency from (intra-) daily to weekly/monthly?
• Is the volume of workload of this activity sufficient to justify up-front investment & license cost?
• Is the process subject to frequent errors?
• Is the process centralized?
• Is the workload for this activity variable, leading to low team productivity ?
• Does the process involve high headcount?
Does it have potential it be robotized? High
Low
Low High
Po
ten
tial
fo
r ro
bo
tiza
tio
n
Potential for offshoring
Retained local • Regulatory obligations
• Client-facing roles
• General oversight over
finance & accounting
Offshored to SSC • Regulatory obligations
• Client-facing roles
• General oversight
Robotization (preferred over offshoring)
• General ledger accounting
• Accounting reconciliations
• Cash disbursement & bank reconciliations
• Fixed asset accounting
• Core business accounting (claims & premiums)
• Data input from other systems (e.g. HR,
operations)
60% 15%
25%
Decision tree for analysis of robotization or
& outsourcing potential
Does it have potential to be offshored to a SSC?
Business & legal constraints to offshoring?
• Is the activity subject to regulation requiring the activity to be performed at local level ?
• Is the activity required to maintain oversight over the accounts & be able to assume legal responsibility?
• Is the activity in strong interaction with the client?
• Is an error in this activity potentially impacting business result?
• Are specific language skills required to carry out the activity?
• Is in-depth personal interaction with other functions required to carry out the activity?
• Is the expertise required to carry out the activity available in the remote location?
"When possible, robotization
should be preferred over
offshoring, for reasons of cost,
quality & speed"
Illustration of segmentation results
Feasibility of robotization?
• Does the activity follow a process that can be largely standardized (vs. subject to many exceptions)?
• Does the activity follow a rule based logic that can be programmed (vs subject to judgment and interpretation)?
• Can the input for the activity be digitalized, in a structured and consistent format ?
RPA case study – Finance function automation C 1
25 20170616 Automation in Financial Services_ADETEM.pptx
Assessment of robotization potential was carried out on a granular list of activities with the local finance teams in the different BUs
Source: Roland Berger project experience
I. Accounting/ reporting General accounting General accounting
I. Accounting/ reporting General accounting Entries to the general ledger (including provisions) Automation / RobotisationFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting General ledger change management Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Cost and revenue allocation principles and systems implementation Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting General accounting Quality assurance, accounting policies, standards proce-dures setting Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Controls over reconciliations Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Booking of reserve setting Retained in local BU Retained in local BU Automation / Robotisation
I. Accounting/ reporting General accounting Booking of impairments of investments Regional Competence CenterFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Booking of tax reserve setting Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing process for full monthly, quarterly, and yearly financial statements Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing Accruals Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Closing Accruals Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing reports for consolidation Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Legal statement for audit, local reports, insource local activitiesRetained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Other Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Investment accounting Investment accounting Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Reinsurance accounting Reinsurance accounting
I. Accounting/ reporting Reinsurance accounting Group reinsurance accounting Automation / RobotisationFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting Reinsurance accounting Reinsurance accounting fronting Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Premiums accounting Premiums accounting
I. Accounting/ reporting Premiums accounting Insurance accounts receivables Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Premiums accounting Billing and charging Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Claims accounting Claims accounting Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts payables
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Fixed assets
I. Accounting/ reporting Non-insurance accounting Fixed assets Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Fixed assets Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Travel & entertainment expensesTravel & entertainment expenses
I. Accounting/ reporting Travel & entertainment expensesT&E booking Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Travel & entertainment expensesT&E payment Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Payroll accounting Payroll accounting
I. Accounting/ reporting Payroll accounting Booking payroll Retained in local BU Retained in local BU Automation / Robotisation
I. Accounting/ reporting Payroll accounting Clearing payroll Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting General accounting
I. Accounting/ reporting General accounting Entries to the general ledger (including provisions) Automation / RobotisationFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting General ledger change management Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Cost and revenue allocation principles and systems implementation Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting General accounting Quality assurance, accounting policies, standards proce-dures setting Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Controls over reconciliations Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Booking of reserve setting Retained in local BU Retained in local BU Automation / Robotisation
I. Accounting/ reporting General accounting Booking of impairments of investments Regional Competence CenterFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Booking of tax reserve setting Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing process for full monthly, quarterly, and yearly financial statements Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing Accruals Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Closing Accruals Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing reports for consolidation Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Legal statement for audit, local reports, insource local activitiesRetained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Other Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Investment accounting Investment accounting Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Reinsurance accounting Reinsurance accounting
I. Accounting/ reporting Reinsurance accounting Group reinsurance accounting Automation / RobotisationFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting Reinsurance accounting Reinsurance accounting fronting Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Premiums accounting Premiums accounting
I. Accounting/ reporting Premiums accounting Insurance accounts receivables Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Premiums accounting Billing and charging Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Claims accounting Claims accounting Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts payables
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Fixed assets
I. Accounting/ reporting Non-insurance accounting Fixed assets Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Fixed assets Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Travel & entertainment expensesTravel & entertainment expenses
I. Accounting/ reporting Travel & entertainment expensesT&E booking Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Travel & entertainment expensesT&E payment Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Payroll accounting Payroll accounting
I. Accounting/ reporting Payroll accounting Booking payroll Retained in local BU Retained in local BU Automation / Robotisation
I. Accounting/ reporting Payroll accounting Clearing payroll Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting General accounting
I. Accounting/ reporting General accounting Entries to the general ledger (including provisions) Automation / RobotisationFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting General ledger change management Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Cost and revenue allocation principles and systems implementation Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting General accounting Quality assurance, accounting policies, standards proce-dures setting Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Controls over reconciliations Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Booking of reserve setting Retained in local BU Retained in local BU Automation / Robotisation
I. Accounting/ reporting General accounting Booking of impairments of investments Regional Competence CenterFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Booking of tax reserve setting Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing process for full monthly, quarterly, and yearly financial statements Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing Accruals Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Closing Accruals Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Closing reports for consolidation Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting General accounting Legal statement for audit, local reports, insource local activitiesRetained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting General accounting Other Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Investment accounting Investment accounting Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Reinsurance accounting Reinsurance accounting
I. Accounting/ reporting Reinsurance accounting Group reinsurance accounting Automation / RobotisationFinance Shared ServiceFinance Shared Service
I. Accounting/ reporting Reinsurance accounting Reinsurance accounting fronting Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Premiums accounting Premiums accounting
I. Accounting/ reporting Premiums accounting Insurance accounts receivables Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Premiums accounting Billing and charging Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Claims accounting Claims accounting Automation / RobotisationRetained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts payables
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts payables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Accounts receivables Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Accounts receivables Retained in local BU Retained in local BU Retained in local BU
I. Accounting/ reporting Non-insurance accounting Fixed assets
I. Accounting/ reporting Non-insurance accounting Fixed assets Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Non-insurance accounting Fixed assets Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Travel & entertainment expensesTravel & entertainment expenses
I. Accounting/ reporting Travel & entertainment expensesT&E booking Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Travel & entertainment expensesT&E payment Finance Shared Service Finance Shared ServiceFinance Shared Service
I. Accounting/ reporting Payroll accounting Payroll accounting
I. Accounting/ reporting Payroll accounting Booking payroll Retained in local BU Retained in local BU Automation / Robotisation
I. Accounting/ reporting Payroll accounting Clearing payroll Retained in local BU Retained in local BU Retained in local BU
Illustration of analyses carried out
Highly detailed mapping of activities and sub-activities (c. 80 activities for c. 125 FTEs)
Determination of FTE baseline per geography
Systematic assessment with local teams of possible levers per activity:
– Offshoring to shared service centers
– Robotization & automation
– Required to be retained in local BU
Interviews with local people to understand drivers & complexity of activities
Cornerstones of the approach
BU 1 BU 2 BU 3
BU 1 BU 2 BU 3
BU 1 BU 2 BU 3
Overall coherency check & further challenge of the allocation results
RPA case study – Finance function automation C 1
26 20170616 Automation in Financial Services_ADETEM.pptx
Through these experiences, we acquired a broad view on robotization potential within Finance functions
Robotization potential across insurance companies – illustration
Large scale robotization Partial robotization No robotization
Source: Project experience, Roland Berger
Finance process
Transac- tional Accounting
Insurance Accounting
General Accounting
System support
Other Finance
Accounts payable + TE
Accounts Receivable
Bank reconciliations/ cash applications
Fixed assets accounting
Re insurance accounting
Premiums accounting
Claims accounting
Investment accounting
Solvency II support
General accounting
Third party accounting
Credit control & reporting
Finance solutions (MDM)
Management reporting
Data analytics
Legal & Compliance Support
Actuarial support
Procurement
Global Corporate insurer
Global service company
Global re-insurer
Large global life insurer
Large national life/non-life insurer (1)
Large national life/non-life insurer (2)
Large national life/non-life insurer (3)
RPA case study – Finance function automation C 1
27 20170616 Automation in Financial Services_ADETEM.pptx
Robotization was often preferred to offshoring – 40% of the accounting team was directly impacted by the project
Achieved results Key learnings
> 40% of accounting organization directly impacted by the project results in a first phase, with additional potential to be further investigated at a later stage
– Reduction of ~25% of total staff through robotization projects, allowing the remaining organization to focus on more value adding tasks
– 15% of total staff relocated to a shared service center, to realize activities not suitable for automation at a lower cost
> Run-rate cost reduction of ~30% compared to overall labor cost of the function, taking into account all costs related to offshoring
> Process to be optimized as much as possible before robotization, to ensure adequate quality level and limit system investments
> High activity volume & frequency are preferred scope for automation to counterbalance the required investments & workload
> Complexity of processes only shows when detailed analysis is carried out
Source: Roland Berger
Return on experience of the project
RPA case study – Finance function automation C 1
28 20170616 Automation in Financial Services_ADETEM.pptx
Watson was implemented on two pilots highlighting savings opportunities of up to 50min/day on account managers by 2020
Productivity savings estimates [2016-2020] – Pilots
Source: Roland Berger
90%
70%
2016 2017 2018 2019 2020
55%49%43%36%
2016 2017 2018 2019 2020
30%
Email analyser
2020 2016
2020 2016
50 min. 25 min.
Detection rate
Total
Virtual assistant
Satisfying answers rate
> Automatized identification of e-mail intent and level of priority, sorting
and visualization based on those two criterias
> Automatic login into IT applications and pre-filling of some information in
the target application
> Customized client answer proposal
> Automatic answers on simple cases
> Machine learning leveraged to continuously improve successful detection rate
Watson Performance Description of levers
Productivity savings [min/day]
> Chat bot to answer simple and recurring questions on products
> Connection to the document database
> Display of a short list of information specifically extracted
– Probability estimate of successful answer
– Link to relevant documents
> Machine learning leveraged to continuously improve successful answer rate
AI case study – Watson assessment and prioritization C 2
29 20170616 Automation in Financial Services_ADETEM.pptx
Illustration – Bottom-up analysis and sizing of e-mail activities Split of activities per profile [hours/day]
Watson extension to new use cases was assessed through a bottom-up analysis of account managers activities
Admin. work
E-mails
Profile C
8,0
0,7
0,3
3,9
0,5
1,4
Profile A
8,0
1,8
1,8
0,5
2,3
0,2 0,5
Profile B
8,0
0,7
Meetings
3,2
0,7
1,9
Meetings prep.
Researches
0,2
Operational tasks 0,9
1,2 1,2
Assessment of Watson potential - Analysis of an account manager typical day
Activity where Watson could prove useful in most areas
Source: Roland Berger
1%1%1%1%1%
1%2%2%2%2%2%
0%
6%6%
7%
3%3%3%3%4%
4%4%4%4%4%
5%
Rendez-vous - Le client souhaite obtenir un rendez-vous avec son chargé de clientèle ou passer en Caisse/en Agence
Contact - Le client souhaite être contacté par le Chargé de clientèle
Editer - Le client souhaite que la banque lui transmette un document
Document - Le client souhaite transmettre un document à la banque
Proposition - Le client souhaite bénéficier d'une offre commerciale de la banque
Ecriture - Le client souhaite des informations sur une ligne d'écriture de son relevé de compte (frais, commissions, etc)
Communication - Le client souhaite informer la banque d'une opération à venir
Virement - Le client souhaite effectuer un virement
Modifier - Le client souhaite modifier un contrat
Souscription - Le client souhaite ouvrir un contrat (assurance, prêt, etc)
Clôturer - Le client souhaite clôturer ou résilier un contrat
Négocier - Le client souhaite négocier une tarification
Moyen de paiement - Le client souhaite savoir si son moyen de paiement est disponible
Personnel - Le client fait part d’un changement d'information le concernant
Lever blocage - Le client demande le déblocage de sa carte bancaire
Tarification - Le client souhaite se faire expliquer une tarification
Renégocier - Le client souhaite renégocier un crédit
Situation – Le client fourni des informations liées à une situation débitrice
Rembourser - Le client souhaite rembourser son crédit par anticipation
Débloquer – Le client souhaite débloquer un crédit
Créancier - Le client souhaite interdire un créancier
Rejeter - Le client demande de rejeter ponctuellement un prélèvement
Chèque de banque - Le client souhaite l’émission d’un chèque de banque
Résiliation – Le client demande à clôturer un virement permanent ou un versement programmé
Fraude – Le client informe sur une fraude le concernant
Opposition – Le client souhaite faire une opposition de sa carte bancaire
Split of different e-mail intentions Difficulty Estimated time [min]
48
46
444
68
106
44
2888
64
26
42
62
4
Activity where Watson could prove useful in some areas
AI case study – Watson assessment and prioritization C 2
Confidential
30 20170616 Automation in Financial Services_ADETEM.pptx
We identified additional AI use cases which could lead to significant productivity savings
Productivity savings estimates on potential extensions [2020] - not exhaustive
> Automation of answers regarding document requests (identification of detailed intent, proposition of answer with documents)
> ~x min / day
> Automation of answers to information requests on fees (identification of the intent, proposition of a standardized answer for frequent cases)
> ~x min / day
E-mail analysis
Description of the levers Productivity savings
> Partial automation of contract modifications or changes in client information (including field matching and manual validation)
> ~x min / day
> Automation of rejected payment requests > ~x min / day
> Automation of meeting preparation: Client history & status synthesis, Product recommendations > ~ x min / day
> Partial automation of meeting minutes: Filling of specific field based on minutes in free text > ~ x min / day
> Automation of overdraft management: Recommended decision (no action /e-mail relaunch/ blockage) and standardized e-mail answers according to client history and situation
> ~ x min / day
Processing assistant
> Extension of the virtual assistant to additional fields
– Financing
– Insurance…
> ~x min / day
Virtual Assistant
> ~ x min / day > Client value management support: Prioritized listing of clients to contact
Source: Roland Berger
Commercial assistant
AI case study – Watson assessment and prioritization C 2
31 20170616 Automation in Financial Services_ADETEM.pptx
Overall, the productivity improvements could reach until 16% in 2020
Source: Analyse Roland Berger
Gains de productivité liés à Watson : Estimation de l'impact ETP en fin d'année [2017 – 2020; ETP 1)]
Gains de productivité (potentiel total) sur le réseau du client [2017-20]
Dec. 2019
Périmètre et évolution d'impact client
Périmètre client, évaluation revue
3 : Extension du périmètre des cas d’usage actuels
4 : Nouveaux cas d’usage
Dec. 2020 Dec. 2018 Dec. 2017
1
2
3
4
Impact en % des chargés de clientèle 4% 10% 13% 16%
AI case study – Watson assessment and prioritization C 2
32 20170616 Automation in Financial Services_ADETEM.pptx
Watson roll-out on account managers network shows potential for 16% of productivity savings
Achieved results Key learnings
> 16% productivity savings on total account managers network
– Revised potential on the pilot scope based on analysis of activities and pilot results
– Extension of existing use cases to additional scopes
– New use cases identified as part of the study
> Progressive ramp-up of productivity improvements over 4 years
> In-depth analysis of activities brings additional insights on AI potential
> Almost all activities can be partially or totally automatized with AI (even interactions / conversations with customers)
> Machine learning gives an advantage to size and experience / AI boundaries can gradually be pushed very far
> 2 types of AI solutions providers : "universal" (eg. Watson) vs. "vertical" (eg. fintechs)
> Social acceptance and impacts of AI solutions to be carefully handled and anticipated
Source: Roland Berger
Lessons learnt
AI case study – Watson assessment and prioritization C 2
33 20170616 Automation in Financial Services_ADETEM.pptx
D. How to move forward?
34 20170616 Automation in Financial Services_ADETEM.pptx
Financial Services players can consider several routes to leverage digital optimization opportunities
Approaches to digital optimization Digital levers
RPA … AI
Pro
cess
es
On-boarding
Mortgage re-financing
…
End-to-end digital re-engineering
> No restriction redesign
> From customers perspective
> Bottom-up "Reality Check"
Sustainable holistic transformation
C
Digital lever maximization
> Systematic across all process steps
> Combination of levers
Focused short term results
A
Marketing
Back-office
….
Digital AZBB
> Comprehensive activity review
> Digital and non-digital levers
Short/Medium-term impact
B
Approach D
35 20170616 Automation in Financial Services_ADETEM.pptx
P&C personal claims
Personal policy admin/underwriting
Risk management
Marketing analytics
Procurement BI
Payments processing
Business banking origination/servicing
Acturial & claims analysis
FP&A Finance MDM
Marketing automation/ campaign mgmt.
Transactional procurement
P&C agency support
Collections
AML Mortage servicing
P&C commercial underwriting
Sourcing/category mgmt.
Supplier risk & perf. mgmt.
Marketing MDM
Procurement MDM
Mortage origination
Basel2
Dodd frank compliance P2P
O2C
Auto finance
KYC/AML
Stress testing
Loan underwriting/origination
Loan portfolio mgmt.
Multi-channel customer mgmt.
P&C commercial claims
Account set-up/servicing
Retail brokerage
R2R
Equipment finance
Basel implementation
Retirement services
Tech
no
log
y ap
plic
abili
ty
Impacting important business challenges
Marketing Procurement Risk Finance Banking & Insurance Operations
Source: Genpact; Roland Berger
Technology applicability and impact on business challenges
Example – Many processes have a potential of using automation techniques but need to address most important challenges first
Few Many
Low
High
Potential automation applications which large impact on important business challenges
Target functions/
processes to automate
Source:
http://www.genpact.com/downloadable-content/insight/the-impact-of-technology-on-business-process-operations.pdf
Approach D
36 20170616 Automation in Financial Services_ADETEM.pptx
Example – An integrated RPA/AI roadmap needs to be designed along with a dedicated Target Operating Model to reach full potential
Illustrative – RPA/AI roll-out
> Understand which areas/processes offer potential for automation, evaluating
– Data intensity
– Complexity
– Volume
– Criticality
– Stability
– Systems and people involved
– etc.
> Prioritize areas of application based on savings potential vs. need for technical and organizational transformation
> Technological transformation
– Understand options, preconditions, and limitations of automation, including dialogue with selected vendors
> Organizational transformation
– Define TOM1) and assess the required changes to enable automation
– Set-up CoE2) : positioning in the organisation, profiles and roles & responsibilities, monitoring and evaluation guidelines
– Evaluate HR implications (personnel transfers, redundancies, organizational re-design, Workers Council involvement, change mgmt., etc.)
> Business plan
– Develop a holistic business plan and transformation roadmap
> Source the technology needed (make, buy, partner)
> Implement successive pilots subsequently enlarging the scope
> Start implementation of organizational measures and change management
> Train employees in new processes
> Measure results
> Evaluate automation in further areas – Implement continuous improvement mindset
Assess full potential Plan transformation Execute plan 1 2 3 > Lower cost
> Higher quality, accuracy, reliability, and compliance
> Focus of workforce on high-value tasks
> Build-up of critical technology know-how for constantly rising digital penetration
Realize benefits
Approach D
Source: Roland Berger
1) Target operating Model 2) Center of Excellence
37 20170616 Automation in Financial Services_ADETEM.pptx
Beyond selected enablers, full and lasting impact of digital levers relies on several sustainability factors
Source: Roland Berger
Implementation enablers and sustainability factors of Digital levers
Implementation enablers @
Cloud computing:
Simple and quick deployment of technologies, limited costs to store/use
large amount of data and to host solutions,…)
Data Management:
Availability and completeness of data, consistency of data quality, clear
governance and ownership over data
Visibility on activities/processes
Critical to prioritize, provides initial material for each lever, enables faster
time-to-market for re-engineering
Enablers Sustainability factors
Robot management
> Adapt the organization to integrate digital management
> Center of Excellence to cover governance (IT & Ops) / technology / Expertise rollout
HR adaptation:
> Fully leverage the downsizing or reprioritization opportunities with impacted teams
> Development of new skills
Compliance evolution
> Necessary to adapt to newly automatized process
> Requirement of new certifications
System orchestration
> Critical to reach end-to-end automation where possible
> Scoping and prioritization
> KPIs used to monitor the orchestration
Approach D