Robotic Process Automation
- 1. Advanced Business Analysis Training Robotic Process Automation (RPA)
- 2. Page 2Classification: Restricted Agenda • What is RPA? • Making Office Productive • Consequences • Automation
- 3. Page 3Classification: Restricted What is RPA? • RPA is robotic process automation. It is not an industrial robotic technology or a physical robot. It is only a nonphysical software robot performing process activities by replacing human intervention or activities with application. There is no much difference in RPA and ITPA(Information technology process automation).After all, they both leverage software tools to automate processes – promoting efficiencies, lowering costs, raising accuracy and elevating quality.
- 5. Page 5Classification: Restricted Advantages of RPA • Codeless • Lower Cost • Accuracy & Quality • Consistency • Improved Analytics • Increased Employee Productivity • Increased Customer Satisfaction • Faster • Reconciliation from Multiple Systems • Versatility • Better IT Support and Management
- 6. Page 6Classification: Restricted Disadvantages • Initial Stage setup • Lack of Skilled resources • Maintenance Cost • Not suitable for dynamically changing tasks • 100% Process automation is impractical
- 7. Page 7Classification: Restricted RPA conception used in Financial Sector Banking: • Report Generation • Approval Process for Mortgages • Fixed Asset Amortization • Foreign Exchange and Bad debt write-offs • New account (multiple) entry • Journal entry
- 8. Page 8Classification: Restricted Continued… Insurance: • Part of Underwriting Services and policy issuance • Electronic filing of policy documents • Premiums comparison • Update customer data • Update data for new products • Recoveries and Payments Comparison Capital Markets • Reconciliation activities • Checking customer records across various systems • Customer on-boarding process • Upstream and downstream data comparison
- 9. Page 9Classification: Restricted Robotics Process Automation and its implementation challenges • The challenges can range from Business (e.g. to provide clear business process steps), to IT (e.g. to provide correct and reliable IT infrastructure with right setup of security policies), to SDLC (to follow correct development methodology and quality standards). • Minimum wait time. • Addressing missed requirements (business steps and rules) as well as ‘current business process optimization’ while moving to robotics. • Testing the solution may also account new set of challenges. • Focus of testing should not derail from testing the robotics solution to testing the underlying applications.
- 10. Page 10Classification: Restricted Making Office Productive The real impact
- 11. Introduction To overcome short-falls in office technology automation, companies have for many years adopted make-shift alternatives... …hard-copy documents that bridge between departments, self- authored office applications and spreadsheets, interns, low-cost outsourcing agencies… Why change now? Contents Consequences What is the business impact of failing to automate the long-tail of office data processing tasks? •The long-tail efficiency gap •Cost to business? •The burden on teams •What roles are affected? •The wasteful office •Risks from shadow data •Fragmented data sources •Data integrity issues •Harnessing data worth •Options facing managers Automation What is Robotic Process Automation and how does it help? •Swivel chair applications •Automation challenges •Traditional IT solutions •Why Robotic Process Automation brings something new •Examples of use •Measuring up •Solution must-haves •Summary of benefits •Next steps
- 12. | 1. Introduction Many things have changed in the world of business over the last decade… ..but a step change improvement in office data-processing efficiency isn’t one of them.
- 13. Workers are different these days… People today blend how they spend their time like never before… when they work, when they choose to learn, play …and socialize. They work on the move, check business emails into the evenings and on weekends. But in return they expect to access their social apps and communities during work time and be permitted to use their own mobile devices at work as at home.
- 14. Valuations have less to do with the physical assets a business has and focus more on soft assets like the size of online communities, the data assets organizations hold… and the ability of a business to convert ‘clicks to cash’. Companies can scale faster, thanks to cloud computing, and reach larger audiences thanks to the worldwide-web and mobile computing. Business is different too…
- 15. Strange then - while all these changes have happened, levels of office automation and productivity haven’t substantively improved. Fill in form s Create spreadsheets to gather, organize, analyze and report financial information Take data from one system, to then re-key into another Build self- authored apps with desktop tools as a ‘first resort’ when IT has no time Aggregate content on PowerPoint slides to share information Use paper forms to bridge between organizations , processes or systems *Since the 1980, productivity in the office has improved by 3% compared to a 75% productivity increase in factories over the same period. Workers still have to…
- 16. Increasing automation is the second most important strategic priority for shared services and global business services (GBS) leaders. Process automation is seen to be more impactful than analytical software and cloud computing. While office productivity has failed to improve in decades, automation remains a top priority for leaders Source: The robots are coming Deloitte Client Survey of Shared Services and GBS clients, February 2016 | When exploring ways to improve automation 71% would expect to first try to leverage their ERP systems, 44% would plan to use bolt-on tools and 13% so far recognize the opportunity to use alternative technologies and methods.
- 17. | 2. Consequences What are the consequences to businesses (and the people that work for them) of the continued operation of sub-optimal data handling and processing tasks in offices?
- 18. A long-tail of demand exists for technology-led solutions that displace the myriad of low-level manual data handling and processing tasks that employers still call on their workforce to fulfil. According to research, around 60% of occupations could have 30% more of their constituent activities automated Translated proportionately in time, it means each member of your team working a 37.5 hour week could be spending as much as 45-hours a month on things they don’t really want to be doing because these tasks take time away from their primary role – and impact on their quality of life. The long-tail efficiency gap 4 5Source: McKinsey & Co., 2016 Report: Four fundamentals of workplace automation; a structured analysis of 200 individual work activities hour sa monthpe rwork er
- 19. Payroll remains the biggest expenditure items for most businesses. Independent research by McKinsey & Co suggests that automation technologies available today, if adopted as they predict, could make a huge impact on workforce productivity. It could save Western economies around $2 trillion in annual wages and impact on over 130 million jobs. The cost to business? Salaries as a percentage of operating expenses: • 52% health care services • 50% for-profit services • 50% educational services • 22% durable goods manufacturing • 22% Construction/mining and oil/gas • 18% retail/wholesale trade Source: U.S. Department of Labo Bureau of Labor Statistics | There are 130 million clerical jobs in Western Europe. Source: Eurostat Automation could save around $2 trillion in annual wages. Source: McKinsey & Co.
- 20. Failing to automate these light-weight data handling and processing tasks is not a victimless crime. Lots of people have to work with data. These mundane tasks put people under pressure to work late, eat at their desks, multi-task… ...and they spend less time on the things that really matter to their employer. The burden on teams |
- 21. It would be easy to assume that sub-optimal automation of data processing only impacts low-level clerical roles but that’s not the case. Whilst these roles typically perform more data entry and processing tasks, the impact of automation goes much further. In fact, analysis by McKinsey & Co. suggests that even CEOs could release up to 20% of their capacity through automation. What roles are affected? C-suite | Middle managers Everyoneelse
- 22. The wasteful office Use of desktop office applications and hard-copy documents is both unproductive and wasteful. • The average spreadsheet or word-processed document is read less than 5-times in its life. • Relying on hard-copy documents to bridge between the weak points in processes increases paper waste. • Use of paper increases demands for storage and space – i.e. filing cabinets, rooms dedicated to archival. • It places further demands on resources – such as having to scan hard-copy documents to electronic files.
- 23. Security of data has become a board- room issue. Use of office documents and self-authored applications results in a proliferation of content in mobile computing devices, memory sticks, desktop drives and intranet content management systems. Recently, this proliferation of so-called shadow data has worsened with the growth in use of mobile and software-as-a- service (SaaS) cloud based applications; many of which are used without the knowledge or consent of IT. Much of this content is invisible to IT administrators who are unaware of the applications, files or data held within them. Cloud security developer Elastica’s vice-president Eric Andrews says the company’s threat team analyzed 63 million documents stored by its customers in 2015, looking for shadow data Risks from shadow data
- 24. • 59% say, that as a consequence of poor information distribution, they miss information that might be valuable to their jobs almost every day because it exists somewhere else in the company and they can’t find it. • 42% say they accidentally use the wrong information at least once a week, and 53 percent said that less than half of the information they receive is valuable. • 45% say gathering information about what other parts of their company are doing is a big challenge. • More than half have to go to numerous sources to compile information • 40% say other parts of the company are not willing to share information, and 36 percent said there is so much information available that it takes a long time toactually find the right piece of data. Fragmented data sources fail managers Source: Accenture web-based survey of 1,009 managers in companies in the United States and United Kingdom with reported annual revenues of more than US$500 million. June 2006 Middle managers spend more than a quarter of their time searching for information necessary to their jobs, and when they do find it, it’s often wrong. Only half of all managers believe their companies do a good job in governing information distribution or have established adequate processes to determine what data each part of an organization needs.
- 25. • Moreover, nearly half of the global respondents said their lack of confidence stems from a lack of information or easy access to data. The findings are puzzling given the emergence of big data techniques, the proliferation of global networks and the sheer processing power contained even in mobile devices. • One reason for the disconnect between big data and decision making, the Harvard researchers found, is that “silos of data, typically imprisoned in customer, financial, or production systems, are frequently inaccessible by individuals outside the functional group.” • In this regard, 43% of survey respondents said important external or internal data was missing, 42% said data was inaccurate or obsolete, and 33% said they “couldn’t process information fast enough.” Data integrity issues Corporate decision makers have major concerns about access to, availability of, and the quality of internal and outside data A survey of 442 business executives around the world found that corporate decision makers have major concerns about access to, availability of, and the quality of internal and outside data. The result is reduced confidence in their decision-
- 26. A study at the end of 2012 by IDC predicted the “digital universe” will reach 40 zettabytes – that’s around 45 trillion gigabytes – by 2020, a 50-fold growth in a decade. BIG DATA and new data visualization tools are creating ‘data driven businesses’ – but organizations may miss-out on the advantages of BIG DATA because of talent shortages. Industry *predicts a short-fall in IT and analytical roles as businesses seek to harness ever more business insights to drive growth. New technologies and automation approaches have the potential to reduce or remove the need for humans in data analytics roles by automating the harvesting, analysis, report formation and distribution tasks that are today fulfilled by people power. Harnessing data worth …Analyze It.. ..Act on What Matter s *McKinsey is already forecasting the US will face a shortage of up to 190,000 data scientists by 2018. Harvest Data
- 27. Without automation alternatives, departmental managers have very limited options to overcome short- falls in their teams. They can: •Hire more staff - but probably not •Approach IT - although they’re already busy •Contract out - but is it practical, affordable? …and so nothing changes. Options facing managers
- 28. | 3.Automation What automation possibilities and options exist? What’s changed from a decade ago?
- 29. • IT experts suggest that organizations have so far managed to automate between 25 to 40% of their workflows today. • A large proportion of those activities yet to be automated require a human go- between to fill gaps between disjointed systems, or interface between people and systems. • These ‘swivel chair applications’ call on humans to work on several systems at the same time to complete a task (e.g. referencing multiple application windows on one screen and populating another). • Such applications have been out of reach for automation by IT – considered either too difficult or too costly to automate using traditional technologies. The swivel chair applications
- 30. Traditional IT solutionsThe procurement approaches of businesses have traditionally favored tool-kit style technology platforms able to perform a common capability across the enterprise. In the context of process automation, in addition to the inherent capabilities found in enterprise resource planning and financial management systems, Business Process Management (BPM) tool-kits have emerged in recent years to augment workflows that span between discrete systems and silos of operation. Whilst BPM tools have become relatively easy to tailor to serve specific use case requirements, they still require the intervention of IT and demand that ways of working are adapted to suit the technologies being introduced. Rarely do tools offer the dexterity of capabilities to fully eradicate the role of the human as the ‘data processor’. Due to over-pressured and under- resourced IT teams, many office automation projects – with dubious Returns on Investments owing to the very small communities of users - drop to the bottom of the IT priority list where so many projects ultimately go to die.
- 31. Automation challenges are not solely about the dexterity of tech-tools. There are many potential inhibitors to change that can derail projects: •Cost of IT and change can exceed the anticipated return when applications may be used by only a handful of users (or potentially just one). • The time and difficulty of capturing and interpretinguse case needs can become an impenetrable barrier impenetrable barrier to projects moving forward. Mistakes made at this stage can cause managers to lose confidence in justifying future projects. • Building a Return-on-Investment (RoI) argument to justify investing IT and project resources can be difficult – if not impossible – when it’s not clear how much time and resource is needed to build the solution, or what value stakeholders will eventually enjoy once the automation is deployed. Automation challenges
- 32. Robotic Process Automation describes not just a new flavor of technology but a new approach to addressing office- centric data handling and processing tasks - those that previously required manual re-keying, spreadsheets, hard- copy documents and swivel chair workers to perform. Why Robotic Process Automation brings something new There are several fundamental characteristics of this kind of technology- enabled solution that differ from Business Process Management (BPM) tooling and other traditional IT department dependent solutions: •The automation does not fundamentally change the operation of the process. Instead of a human tapping keys and building spreadsheets, a computer software application mimics the same tasks in a similar way. Stakeholders do not see a fundamental change to their operating behaviours. The same process stages will generally occur. • Training of robots is generally performed using code-free software, so it’s quick to do and easy to adapt. • IT challenges like implementing complex APIs are removed by re- keying or robotically extracting/uploading data to legacy systems. • Rather than being ‘a technology platform’ purchased and operated by IT – requiring a big up-front investment and lots of risk, ‘robots’ are a resource that can be outsourced in the same way as people, individually selected by departmental managers and billed monthly – with all the contractual flexibility that represents.
- 33. Robots have different skill-sets, just like humans • Check • Capture (multi- sources) • File • Share (Email) • Transform • Route • Data Input • Record • Artificial Intelligence learning engine • Non repeatable One off Linear tasks manual tasks• Standard • Repeatab le Orchestrat ed activities • Complex • Standard • Multi- scripted • Validation • Compare • Calculate • Key-fill • Analytics • Security Trace Dynami c process es • Non-standard • Contextual • Inference • Autonomic deployment and control • ERP & Legacy systems APIs • Formatted Omni- Channel Communications Systems • Predictive • Self Learning • Self Healing AttributesSkills Orchestration Autonomics Cognitive
- 34. Examples of use The global IT robotic automation market was valued at US$ 183.1m in 2013 and is forecast to grow at 60.5% CAGR from 2014 to 2020. Source: Transparency Market Research
- 35. What does the robot do? Harvests data from motor dealership back-office systems to identify sales lead opportunities. It then publishes the sales leads into the dealerships (separate) customer relationship management system; delivering high quality leads, one lead at a time. What alternatives were considered? •Generate a one-off monthly report (this was trialed but, when faced with a long list of unqualified leads, sales people would not act on the information) •Do nothing •Pay an intern or administrator to manually capture the data and upload it Automotive Sales Assistant Benefits: •Data accuracy •Centrally managed •Resources were not available in-house to support this process •The legacy system used meant generating reports required lots of manual data aggregation and cleansing before leads could be forwarded. •Leads needed to be manually entered into the CRM system. •Information relating to the sales opportunity useful to the salesperson could not be harvested from one system, so the usefulness of leads generated with a robot would be of little use for salespeople. Examp le Use Case
- 36. What does the robot do? To validate requests to perform road works, a robot will harvest data from multiple sources and map assets to build up a complete picture of the context of the road work to be undertaken compared to other local traffic planning events. What alternatives were considered? •Do nothing and continue to use an administrator, although the traffic manager knew that growth in requests meant the team would be short-handed. •Purchase a new works planning system, but the budget and intrusiveness of a new system was not desirable. Road Works Admin. Assistant Benefits: •Systematic; rules are applied consistently •This was an automated a swivel chair application that removes the need to recruit and train additional member(s) of staff •Removes resource planning challenges within department (covering lunch-breaks etc.) •Lower investment requirement •As the user organization is governmental, easier to justify expenditure as ‘best value’ approach •Easily and quickly scalable •Easily repeatable Examp le Use Case
- 37. Robotic process automation is best suited to mundane clerical data handling, re- purposing, re-keying or re- publishing activities that: •Can be mimicked by technologies that are script- based, use if-then conditional logic or employ artificial intelligence •Tasks that involve small numbers of users (potentially just one) that would be uneconomic to automate with traditional IT •Tasks that do not vary every time they occur Measuring up
- 38. • Doesn’t require coding or developers to ‘train’ – which might result in a need for testing, tuning, de-bugging etc. • Adopt de facto industry standards to minimize/remove information security risks, scaling and performance tuning overheads etc. • Is scalable and flexible • Can harvest and use data from multiple sources at the same time (multi-threaded, multi-sourcing) and potentially in real- time – including sensors, CCTV image or video feeds, DLLs, web portals, GPS and web apps Solution must- haves Applications design is a wasteful process. The article ‘Why Your IT Project May Be Riskier Than You Think’ by HBR (Nov 2011) followed a survey of 1,471 IT projects with an average spend of $167m and found: •The average overrun was 27% •One in six of the projects studied was a ‘black swan’, with a cost over-run of 200%. •Almost 70% of black swan projects also over- ran their schedules. Robotic Process Automation removes project over-run risk and delivers best-fit solutions.
- 39. 1. Releases time to key workers - Performs mundane clerical tasks so key talent can focus on those activities that bring value to customers and stakeholders. 2. Stays in-tune with business tempo – Much faster to build and deploy than traditional solutions thanks to its light touch codeless approach (measured in days and weeks rather than months) all but displacing development and IT skills. 3. Keeps working 24/7 – Robots can work day and night on their designated tasks to maximize the ROI available. 4. Truly agile - Solutions can be re-trained as needed must be adaptable to a variety of business needs and scalable to enterprise size. In addition, solutions must be compliance-ready and secure, storing nothing locally. Summary of benefits 5. Eliminates Human Errors – Automation with RPA eliminates human error; fulfilling processes predictably every time. 6. Scalable – Most robotic process automation platforms are designed for enterprise scale deployment and are typically cloud based permitting truly massive scaling at very low cost. 7. Self-serving – Advanced solutions offer user organizations the added benefit of being able to train their own robots! 8. Easily integrated - Complementing, rather than replacing, existing systems RPA solutions can painlessly access data from multiple, disparate sources without needing APIs or coding.
- 40. • Robotic Process Automation is not for everyone. Some organizations have re- designed their business models to completely eradicate data handling and processing. • But, if you find people are spending too much time re- keying data, printing or using spreadsheets, then the best thing you can do is to arrange an RPA consultant to run a diagnostic audit across your process flows. Next steps Top tip: To consid er if RPA is right for your organi zation, find out what people use paper forms and office
- 41. Page 41Classification: Restricted How are you affected Try exploring
- 42. Page 42Classification: Restricted What is RPA? • Leave apart fancy name, I would say RPA is a simple way to automate Business Processes without building new IT Systems and without re- engineering the processes. It is becoming popular because of the ease of automation without process re-engineering. • The way RPA works is, tool can record your actions and replay those actions. Till the time your actions are repetitive, you can automate the process. With time RPA tools have become sophisticated - integrating with various IT systems like Databases, Excel, SAP etc.
- 43. Page 43Classification: Restricted How this is different from Normal Automation? Software robot simulate manual clicks with records actions and business logic. Normal automation requires new systems to be built for automation; Normal automation is usually costly and time consuming. Normal way to automate: • Develop new enterprise system: This is usually most complex - requires high time and high cost. • Optimize the process with Business Process Management Systems: This usually may be less complex than developing new IT systems but costs & time required are almost same as developing new enterprise system RPA is much easier than both above mentioned automation approaches. It doesn't require one to re-engineer process or build new IT system.
- 44. Page 44Classification: Restricted What type of processes can be automated with RPA? Repeatable, simple, standardized IT system interactions by human users are low hanging fruits for RPA. Till the time a process can be written as workflow, it can be automated with RPA. Following are commonly used applications by users. Most of RPA tools supports recording actions on these applications and which can be replayed as automation: • Web based • Desktop • Citrix • SAP • Excel
- 45. Page 45Classification: Restricted Are Software Robots better than Human? Software Robots are not to replace human but to help human workers. Software robots are made for processes which are mundane, repetitive and not liked by human most human workers. For mundane, repetitive tasks which are in scope for RPA, Robots are certainly better than Human workers: • Robots can run 24*7 • Robot behaviour is predicatbale • Robot brings auditable logs
- 46. Page 46Classification: Restricted How can it impact you? Well, you might have guessed by now, if some part of your job requires you to do mundane, repetitive tasks - Robots are here to help you. But if your job requires you to just do these mundane, repetitive taks then you must start reskilling yourself do better job.