Blue Prism: How to maximize Business Value and Return on Investment
Neil Wright, Executive Vice President – Professional Services, Blue Prism discusses the harmonisation between IT and the Operations Functions.
By 2030, a major portion of ERP-related work may be handled by machines. These systems will increase in capability as the amount of data grows and as AI advances. Human-machine interactions will play a major role in business, and well before then.
The importance of human-machine interactions to business was ranked very high by the respondents participating in research by Dell Technologies and the Institute for the Future. The report is based on a survey of nearly 4,000 business leaders. More than eight in 10 (82%) overwhelmingly agreed that they “expect humans and machines will work as integrated teams within their organization inside of five years.”
Further out, by 2030, smart machines will play an important role in ERP. Three of the top four functions that will be offloaded to machines, this survey found, are ERP-related: inventory management, financial administration — invoicing, purchasing orders, etc. — and, in fourth place, logistics. Troubleshooting was number three.
But overall, there is a lot of uncertainty about the technological future.
When asked if “automated systems will free up our time,” the response was split down the middle, with half agreeing and the other half disagreeing.
In an interview, Danny Cobb, Dell Technologies corporate fellow and vice president of global technology strategy, discussed human-machine interactions and other survey findings.
Cobb sees a wide range of qualitative and quantitative processes and technologies — AI, context and pattern recognition, voice and image recognition — gaining enterprise use. His responses were excerpted and edited.
In 2030, more and more tasks will be offloaded to machines. Three of the top four are ERP-related: inventory management, financial administration and logistics. What does this mean?
Danny Cobb: It’s hard to imagine that that’s the first thing someone thinks about [inventory management, financial administration] in their digital transformation agenda, but it also paints a picture: We’re not as far along or sophisticated as we may think we are if those are still some of the topics that come up.
Does this mean that things like inventory management will be more automated? That something like image recognition might be used to track product as it moves through the supply chain?
Cobb: That’s right. You see image recognition, drone technology and robotic technology assisting with that function. You see maybe more global logistics functions that might be operating in a hybrid cloud or a multi-cloud way that gives a broader insight into all the inventory and material capability of an enterprise, 24/7 and around the globe.
ERP systems will be handling a lot more data from a much wider range of sources. What do those systems begin to look like in the future?
Danny CobbCorporate fellow and VP of global technology strategy, Dell Technologies
Cobb: At the edge of an enterprise — the edge being wherever the first unit of intelligence begins to exist — there might be a stream of telemetry. It might be all this inventory data. It might be all the input from these drones, or from a global logistics system, or from multiple systems because of supplier-to-supplier linkages. These systems now need to be much more intricately linked than ever before. There is an opportunity for an entirely new platform to come into existence — the intelligent edge of the enterprise that handles this telemetry, that handles any of the immediate compute or storage needs. It takes that information and shares it appropriately with a core data center that might contain additional intelligence from the rest of the enterprise. The edge technologies do the first stage of work, and then, those migrate upstream to a set of core technologies that are responsible for further analysis, long-term storage or broader distribution.
How much data will we be getting from these alternative sources, and what are the challenges to processing it?
Cobb: Artificial intelligence, machine learning sorts of capabilities are going mainstream because the amount of compute that we have has caught up or is catching up with the amount of useful data that’s there to be analyzed. Other instrumented systems [such as autonomous vehicles, building automation systems, jet engines] are throwing off a tremendous amount of data, and we can now afford to process it as it’s being generated. We can now embed processing in just about anything.
A high percentage of those surveyed for this report expect to see more human-machine interactions by 2030. What does that mean?
Cobb: It may not be strictly a physical presence — a personal robot sitting in that room with me — but artificial intelligence itself will complement the team’s function and will provide a useful value. It’s that sort of digital partnership.
How useful is it to think about the world 15 years or so from today?
Cobb: They [customers, users] need to start getting a blueprint that helps them address some of these opportunities or manage some of these risks. What research like this does is to give customers a vehicle for thinking about this. What are the new roles that are going to be created? What are the skill sets that need to come into existence? How might that impact job satisfaction?
Be prepared when implementing Robotic Process Automation. Learn about the most common points of failure when automating business processes.
The principle that technology performs technology jobs and humans perform human jobs is a simple one. Virtual and human workers each have their own strengths that should be employed and valued. The difficulty lies in determining the best fit for each role and responsibility, especially when a lack of standardization is involved.
Robotic process automation (RPA) is large part of future-of-work technology. To use RPA optimally, standardization is required. The fewer steps an automated solution performs, the quicker it will run. Additionally, if a single set of instructions exists, a single solution can be built. If highly variable instances with many different requirements exist, multiple solutions must be designed and built.
What Is a Standardized File?
Most clients understand that having structured data (e.g., a field-based file) that uses a good data type (text and numbers) is vital for RPA. But there’s a common misconception that having the same data on a page, regardless of position or format, is equivalent to standardization. To be fully standardized, the data must appear in exactly the same cell, field or position in every instance.
The same goes for the processes. It isn’t enough that certain actions take place. Every instance of each action must be performed in exactly the same way, in the same order, using the same rules.
What Are the Benefits of Standardization?
The following are some benefits of process standardization.
- Simple, well-defined standard operating procedures (SOPs). When you’re using highly standardized processes, you can write simple and well-defined SOPs that exclude futile work and bad practices. Employees are less likely to develop individual workarounds. Having clear SOPs is also useful for compliance.
- Ease of training new employees. Clear and simple rules allow quick and structured training for new employees. They also mean that when associates have learned how to interact with and process one client, they can work with all clients. This benefit is especially useful when work must be transferred from one employee to another for reasons such as annual leave or attrition.
- Ease of adding new clients. You can take on new clients with ease, as all new clients undergo the same process. In closing the sale, clients will be aware of exactly what input they must deliver and the business can be transparent with them from the start.
- Increased scope for RPA configuration. Any simplification and process re-engineering that occurs before RPA development will make the deployment faster (meaning the benefits will accrue sooner). The RPA solution won’t need reconfiguration for new clients, and it’s more adaptable for future process developments.
How Can Standardized Processes Help Meet Individual Client Needs?
It’s important to ensure clients feel well served and unique, and some varying requirements between clients will be inevitable. Almost all processes, however, provide opportunities for standardization. For example, legislative parts of processes tend to be common across all clients. If they are standardized and then deemed suitable for RPA, associates will have more time to spend meeting the bespoke requirements of your clients.
For example, in one of Symphony’s completed future-of-work Assessments, a process involved receiving fresh information from the customer each month. The employees would create Excel templates for each customer, but only about 30 percent of the customers submitted their details using the template. Other data arrived in a modified version of the template; a customer-created Excel file; a Word, PDF or alternate application file; or a free-form email that attached a mixture of the above. Not only was the associates’ effort wasted in creating the template, but dealing with the mixed-data input was a difficult and time-consuming task. By analyzing the business case they discovered that if the clients enforced template usage and increased their standardization levels from 0.3 to 0.6 (over all processes) before implementing RPA, their three-year ROI jumped from 300 percent to over 450 percent.
In conclusion, standardized processes should be the same in every instance. The benefits of having standard processes apply to the workflow, the employees, the clients, RPA development and finances. As such, process re-engineering to increase standardization is critical for an organization’s journey into the future of work.
It’s everywhere. It’s all around you. It’s in your smartphone, in your e-mail, in your Amazon and your Netflix, in your car and in your favorite supermarket. It’s in Google’s CAPTCHA, in the stock market and probably behind the recent presidential vote. It’s in genomic sequencing, in particle physics and astronomy. What is it?
It’s Machine Learning. And it’s changing the world as we know it.
You give me data, I give you (instant) gratification
Here’s a question: why would someone ever want to keep in their house a machine that collects information about them 24/7 for purposes that are arguably beyond their knowledge and control?
It’s a trade. You entrust me with your data, and in return I give you answers to your questions, product recommendations and dating suggestions tailored to your interests, optimized driving routes, spam filters, or a new credit card.
As our digital footprint deepens, most of the data we continuously generate is being collected, processed and transformed into useful products or services. Just as Google’s algorithms determine to a great extent what information you find, Amazon can largely influence what products you buy.
Machine Learning (ML) algorithms have an extraordinary capacity to process vast amounts of data and find patterns in it. And the more data there is, the more they learn. For many applications—from vision to speech to robotics, and in different areas of business—from retail to finance to manufacturing, Machine Learning is becoming the new driving force.
To give you a rough knowledge of this technology, a conceptual model to better navigate the expert field currently taking our own industry, automation, to new heights, today we’ll introduce you to UiPath’s team of Machine Learning developers. Stefan Adam, Virgil Tudor, and Dragos Bobolea are the geeks who are leading the research and development of Machine Learning here at UiPath.
Guys, what is Machine Learning?
Virgil: Machine Learning is a subfield of Artificial Intelligence (AI) that enables systems to learn from data. It has at its core Deep Neural Networks, as does most of the current state of the art AI.
Dragos: Deep Learning—the part of ML that we are using—focuses exclusively on multi-level Neural Networks. Basically it involves a network of information that takes pieces of knowledge, combines them in various ways, and finally builds them up towards sensible, high-level meaning.
In the past, AI was composed of lots of very specific algorithms invented for all sorts of problems, from finding contours in pictures to very specialized things like detecting faces. A big part of the job was engineering all this domain knowledge into the algorithms.
Now, thanks to recent advancements in Neural Networks research and hardware computing power, it became feasible to leave this reverse engineering task to a Neural Network and assist its learning process in various ways. The biggest advantage is that Neural Network training, like pedagogy if you will, is almost universally transferable across domains. For example, teaching maths is not that different from teaching chemistry (same teaching method, different curriculum). Similarly, here at UiPath we can use the same state-of-the-art methods that others use for OCR engines, speech recognition, self-driving cars, etc.
Gartner predicts that Machine Learning will reach mainstream adoption in two to five years from now:
“Machine Learning is one of the hottest concepts in technology at the moment, given its extensive range of effects on business. A sub-branch of Machine Learning, called Deep Learning, which involves Deep Neural Nets, is receiving additional attention because it harnesses cognitive domains that were previously the exclusive territory of humans: image recognition, text understanding and audio recognition.”
So what is currently embedded in our Platform in terms of Machine Learning?
Stefan: So actually in the product we have integrated different OCR components. We are using OpenCV to process images, and we also support text analysis based on Microsoft, Google and IBM components. Our image recognition engine uses powerful algorithms that are optimized to find images on screen in under 100 milliseconds. This makes it possible to automate even the most complex applications, available through Citrix and other virtual environments. In fact, it takes almost the same amount of time to build an integration that involves Citrix as it takes to automate a regular desktop application.
And what are we planning to develop going forward?
Stefan: There are three main directions. The first one is related to the way UiPath interacts with the target application—the application which we are trying to automate.
The current detection engine is based on different Accessible API’s. That’s why our screen scraping engine is strongly connected with the execution environment. We plan to incorporate ML especially Deep Learning in our product such that the system will be able to understand any screen, similar to the way humans can understand it. In this way our core detection engine will become invariant to the execution platform. This will also lead to the ability to continuously train our engine by assisting a human user.
The second direction is to offer more Cognitive activities related to natural language parsing and image processing.
And the third direction is to also offer businesses the possibility to build, train and customize different Machine Learning models for performing different tasks, mainly classification and detection.
So with all these enhancements, automation will gradually come closer to emulating and augmenting the power of the human brain.
Stefan: Yes. It has always been the specialty of humans to read and listen to words or capture images. But with the advent of Machine Learning, Natural Language Processing, Neural Networks, Deep Learning and so forth, being able to read text, understand voice and recognize images is also becoming the domain of machines. And the application of these technologies in business will open up possibilities that were previously unimagined.
One of the first, most effective outcomes of applying Machine Learning to RPA will be a newly gained ability of robots to handle complex processing exceptions autonomously. By learning from historical data, they could predict exceptions and prevent anomalies, eliminating the time, effort and cost needed to handle them. All of this will greatly extend the scope of automation to include many activities that involve human judgement.
Using learning algorithms, an RPA robot could make processing decisions contextually, while considering millions of data points from past experience and delivering more accurate predictions. In a claims processing scenario, for example, the robot would automatically review the claim file, eliminate duplicate entries, assess eligibility and then deliver adjudication decisions with human-level precision.
What sparked your interest for this domain?
“(…)In most of the computer science subfields, a scientist or a programmer without basic AI knowledge will be like a blind painter.”
Virgil: The raw power of Deep Neural Networks, the fact that it employs a lot of math and because it’s a mandatory skill for any future computer scientist. In most of the computer science subfields, a scientist or a programmer without basic AI knowledge will be like a blind painter.
Dragos: I guess the thing that awed me the most was this framework of representing knowledge. As a layman, seemingly trivial concepts like “what is a pen” were very fuzzy and ungraspable when I tried vizualizing them. There’s no obvious way of quantitatively representing a pen, so you could say, “look, this picture’s of a pen because this or that.”
Trying to make sense when your pen is just an array of numbers is even more mind-boggling. You start imagining various rules, that get very complicated very quickly, you get scared and think how could anybody ever do this? But that’s what people actually did for a long time (and are interesting on their own). So naturally, I got very excited at discovering methodical ways that can attack these sorts of problems. Also, it’s amazing that we live in a time when we can put them to good use!
The hottest job in Silicon Valley
According to Tim O’Reilly, data scientist is the sexiest job today. Machine Learning experts are rare, forming an elite category that is frenetically being hunted by the big players the likes of Google, Facebook or Amazon. The McKinsey Global Institute estimates that:
“There will be a shortage of talent necessary for organizations to take advantage of big data. By 2018, the United States alone could face a shortage of 140,000 to 190,000 people with deep analytical skills as well as 1.5 million managers and analysts with the know-how to use the analysis of big data to make effective decisions.”
They say Machine Learning is the ideal occupation, because learning algorithms do all the work but let you take all the credit. What have you got to say in your defense?
“The whole concept of learning needs a function that will measure how good or bad a prediction is. If you see a cat, it’s wrong to say it’s a dog. It’s very wrong to say it’s a truck.(…)”
Dragos: A 7 year old knows how to read, right? Imagine giving him a full fridge, and a cookbook. His beef wellington will be similar to the results of an off-the-shelf Neural Network that somebody threw data at. There are at least 3 important things that a Machine Learning developer does:
First, he has to know how different algorithms work, and why they work, so that he knows what tool to choose for the job. Second, he has to figure out how to make the best use of domain knowledge, and give possible “shortcuts” to the Neural Net—this can turn potentially unworkable problems feasible, because it heavily trims down the “number” of bad tries. This is both a problem understanding challenge, and also a technical challenge—you have to write good code for it. Third, once all the pieces are in place, you have to attend to the whole training process, because there are many more ways for it to go on a wrong path than not:
- The whole concept of learning needs a function that will measure how good or bad a prediction is. If you see a cat, it’s wrong to say it’s a dog. It’s very wrong to say it’s a truck. This function—the cost function—turns out to be very hard to design, as it depends very much on the problem at hand (eg. the data distribution).
- You have to find good learning rates at different stages, so that it doesn’t start out too slow, and so that it can learn fine details later on.
- Make sure the model does not overfit the learning data, so that it will perform well in real-life situations.
- Figure out edge cases, adversarial examples, and understand why they occur and how to protect against them.
As a final note, could you share your favorite resource on ML knowledge for all aficionados out there?
Dragos: This is an emerging field, with advancements being made every few weeks. Arxiv.org is the website where most of the research is published, and that’s what you’ll usually find the team sipping their coffee over. For maths and other bottom-up knowledge I love OCW and Stanford’s online courses. And of course, like any developer, we all have that Chrome window with 20 tabs of StackOverflow and Github threads.
Where does RPA currently stand?
Worldwide use of robotic process automation has led to significant, positive impacts on business productivity. In 2017 the adoption of robotic process automation grew globally at a higher pace than ever before.
According to Deloitte, RPA has become a “top priority for global business service leaders” across the industry landscape. Sectors such as banking, insurance and financial services have led the charge with their uptake of RPA to date. Pleasingly, the range of industries now investing in or actively investigating robotic process automation is growing, including manufacturing, utilities, mining, hospitality and FMCG, to name a few.
With its non-intrusive and flexible architecture, an automation scope that encompasses such a wide range of applications and business processes, and its ability to facilitate better management of the changing labour market, robotic process automation has become more than merely a viable option.
Indeed, extrapolating RPA’s growth trajectory into the near future would imply adoption as a necessary step that competitive businesses must take. According to the Institute for Robotic Process Automation and Artificial Intelligence Survey from June 2017, 66% of respondents were considering expansion of robotic process automation programs and 70% were allocating increased funds for investment in 2018.
Zoom in on Australia
According to Frederic Giron, research director at Forrester – quoted by Beverley Head for ComputerWeekly – one of the most ardent concerns of Australian businesses with respect to automation will be innovation.
To this end, companies will focus on sourcing bright (human) minds who can fuse the best of both worlds – human intelligence and collective experience, with the power of process automation. Such automation is expected to work hand-in-hand with broader company mandates to provide personalised products and services that provide the best possible customer experience.
In CiGen’s feature in the Australian Financial Review, Leigh Pullen, Executive Director and co-founder says of the Australian landscape: “RPA has moved to proven technology and is providing tangible benefits to the companies that deploy it”.
Let us now list some of the robotic process automation best practices that can help to lock in these tangible benefits.
Top 6 robotic process automation best practices
1. Select the processes to be automated wisely
CiGen’s Leigh Pullen states that in order to tap into the vast potential of robotics, it is essential to begin an RPA implementation with those business processes that are best suited for automation. Selection should be made based on criteria such as:
- Mature, stable processes – processes that are stable, predictable and well documented, with operation costs that are consistent and well defined
- Low exception rates – low rate of variable outcomes that would otherwise require complex human intervention
- Measurable savings – cost savings and/or benefits gained can typically be expressed in terms of greater accuracy, faster response times, lower cost base, etc.
- High volume / high frequency – these processes often provide a faster ROI
2. Understand the human resources required to build your automation projects
People are at the heart of any successful, sustainable robotic process automation program. There is no “one size fits all” approach to building an RPA team, so taking the time to evaluate your options is key. Will you look to build your team from internal talent, who may have an existing knowledge of business processes and ideal automation opportunities?
Do you prefer to hit the ground running by working with a specialised RPA consultancy, who can build, implement and manage the automation projects for you? Or perhaps a blend of both, where your internal Centre of Excellence is initially guided and trained by an RPA consultancy, before taking full control themselves? Answering these questions at the beginning of your automation journey will help drive a more successful outcome.
3. Have an “RPA sponsor” with a holistic, centralised vision of the road towards RPA implementation
It is crucial that automated processes remain compatible with all the other procedures of your business. Merging of RPA into the overall functionality is therefore a must. This is why having someone who can have a “bird’s eye view” over all the relevant aspects facilitates efficient implementation on a larger scale.
PlantAutomation-technology and ETtech take this one step further and recommend that such a person be an executive sponsor, or someone who can also handle the financial aspects of implementation. Relatedly, an RPA centre of excellence (COE) might ensure the right level of centralisation, which is likely to provide not just short-term process automation but also a coherent longer-term plan.
4. “Divide and conquer”
This piece of advice comes courtesy of Sreyans Jain, Wipro BPS. Suppose after a thorough analysis, you are faced with the reality that the process whose automation is likely to be most useful is also a very complex one. Should you go for the second-best option? Not at all.
It may be better to invest time into breaking this complex process into several sub-processes, and then implement automation only for those that do not require human-level decision making. For instance, the parts that simply gather the information needed to make a decision could be automated, and leave the decision-making itself to human employees.
5. Having a test plan and a fallback plan is critical before deploying an automated process to production
As with any automation technology, rigorous testing of an automation build is critical. It is recommended practice to be prepared with multiple environments where the build can be developed and tested, before finally being approved and deployed. We highlight some important points to consider in order to minimise any frustrations when transitioning to live production:
- For all applications interacted with during the automation, ensure an application version match between Test/Development and Production environments
- Sufficient test data should be made available in Test/Development environments, with the test data set being as recent as possible
- An agreed fall back plan is in place should the automation build require re-work after deployment to Production
6. Train and educate your employees so that they can move beyond the mythology of “robots will steal our jobs”
Never miss an opportunity to explain to your employees what your automation projects are being used for, identify where the automations will assist them in their day-to-day responsibilities and how this forges a pathway towards higher value, business building projects that will tap into the true skills and expertise of your workforce.
Be very clear regarding what robotic process automation can do (and what it cannot), so that expectations are maintained at a realistic level. In doing this, you build and foster a sustainable, long-term development of RPA within the business.
Sreyans Jain makes the enthusiastic statement that “Sooner than later robots will be embraced by all.” We believe that this is the natural conclusion to be drawn, after showing how accessible software robots are in fact. If this information is corroborated with knowledge about the numerous benefits of RPA technology, one cannot but embrace Jain’s optimism about the future of robotic process automation.
I’ve been selling RPA – which includes surface integration, robotic desktop automation (RDA), user interface (UI) optimization, and robotic process automation (RPA) – for over 30 years, and I’ve seen it revolutionize the way a company does business, but only when incorporated into a properly designed system. I’ve seen organizations get many times a purchase’s value in return by using RPA to bridge the gap between their systems, but I’ve also seen the opposite. Too many companies bought RPA software and incorporated it into their systems without significantly re-engineering their business process enough to receive the full benefit, despite my efforts and their internal champions’ efforts.
What Are They Missing?
Today, integration, case management, process reengineering, and newer advanced technologies like natural language processing are coming together to deliver new levels of strategic automation. RPA is a part of this growing toolset, but by itself, it will not ever deliver “end-to-end” transformation. It’s very good at the tactical replication of “manual” human work against older legacy systems. Just as an automatic car is not the same as an autonomously driving intelligent vehicle, RPA by itself is not end-to-end automation.
In fact, the word automation means something significantly bigger. It means that people are using technology to deliver transformation and simplification. Demand for change is coming in faster than ever. Real change now comes from the consumption side (either B2-B or B2-C). Buyers want to engage differently, and enterprises must change to react (and, hence, compete) with how their customers want to transact, which is at any time and on any device.
Use a Strategy that Reworks Your Workflow
Strategic automation technologies are growing at a faster pace today than tactical ones. Did you know you can start building and maintaining new strategic applications significantly faster and cheaper than you’ve ever been able to do so, often in days or weeks? You can build end-to-end journeys and processes without coding by using model-driven platforms that scale to meet IT needs while allowing for business-driven change. That spells freedom for so many organizations seeing complexity in their tools rocket their agility and usability to obsolescence.
Today, you can build and deploy on your choice of platform, be it a public cloud, private cloud, or mixed (or change your mind and simply switch later). You can deliver applications to any current or future device without even thinking about it, even engaging with new channels like chatbots and virtual assistants. Within this single platform, you have AI built in, and it’s making real-time recommendations for the Next Best Action at every interaction point, often even before your customer reaches out to you.
Wrapping AI in front or around RPA does not make RPA any more intelligent. In fact, most RPA advertised as AI or “cognitive” is simply just a repackaging of ever-evolving Optical Character Recognition (OCR) technologies. Know your market. Who actually delivers on promises today with real, functioning technology?
Make Sure You Really Do Tap RPA’s Power
RPA is powerful. It can replicate repetitive, manual tasks to make an employee’s experience better and more productive. Using AI as a feed to RPA by allowing robots to consume unstructured and structured data is a very real and very good thing. AI, machine learning, and intelligent OCR are all things that can help drive more automation of old manual tasks through RPA, but fundamentally you’re still just automating the same old manual tasks. You’re repaving the cowpath with increasingly better tar, but you aren’t building a new superhighway. However, redesigning experiences by embedding AI at the heart of every customer interaction to drive real-time change will have a more significant impact on your business over the medium and long term.
Use RPA to accelerate your transformation, to connect systems that are otherwise not connectable, or to make your employees happier and more productive. But please do not believe that RPA is your endpoint for transformation. The real work is much harder — and the impacts far more significant.
For years, companies have been looking for ways to reduce the cost and burden of routine and repetitive tasks. Many turned to outsourcing and offshoring as a way to accomplish this goal, but outsourcing comes with a whole new set of issues: political, economic and cultural.
Luckily, robotic process automation (RPA) has emerged as a new technology to help free up people for more strategic and fulfilling tasks while RPA handles the routine. However, many people don’t understand or feel comfortable with the use of robots outside of manufacturing, so executives looking to adopt this technology should approach the changeover with care. Here is a checklist to help executives prepare the organization for the use of robots to automate tasks.
A Checklist for RPA
Getting an organization ready for RPA includes:
1. Educating Leadership and Stakeholders
The first step is to ensure that the entire team understands the benefits of RPA and how it can help to streamline processes and control costs. Rather than spend their days doing the same repetitive tasks, stakeholders will now be free to spend their time handling exceptions or engaged in tasks that require more judgment and independent thinking. People think of robotic arms on the shop floor, or androids such as C3PO or Rosie from the Jetsons when they think of robots. RPA is not about physical robots who jump to obey every command. Instead, the focus is on simplifying business processes by developing and automating rules so that most process steps are completed without human intervention.
2. Setting a Future of Work Vision
Just as the use of automation on the shop floor freed workers from much of the drudgery of the assembly line so they could have more autonomy to learn additional skills or take steps to improve product quality, RPA will do the same for office workers and other people involved in repetitive or routine tasks. Instead of spending their days performing the same endless tasks, people will now spend their time handling exceptions or working to provide better customer service, higher quality or new product ideas. The work will be more engaging and interesting for the people, because robots will automatically process the bulk of transactions and procedural steps that fit the norm and alert the worker of exceptions that must be handled manually. Take the time to ensure that the organization buys into the vision, because it is crucial to the success of the project.
3. Process Documentation
Once the organization has bought in to the benefits of RPA, it is time to document existing process steps. Flow charts or other process improvement techniques can help make it clear where the process can be simplified. If processes are not stable, take the time to stabilize before you start to automate.
4. Preparing IT
IT’s role will change greatly with RPA, since many RPA tools are simple enough for skilled end users to use. Rather than relying on the technology skills of the past, IT will find that their role also becomes more strategic, helping to identify processes ripe for automation. In addition, they will work with IT at customers and suppliers to enable collaboration and communication for processes that cross organizational boundaries.
5. Bring in Outside Expertise
Since RPA is an emerging technology, it is imperative to work with people who have the skills and knowledge to help guide you through the process.
6. Identify High Priority Targets
After reviewing your process documentation, you will see some high volume or extremely repetitive procedures that are good targets for the initial foray into RPA. Avoid starting with a critical business process until you have more experience with RPA.
7. Build a Business Case
If you are outsourcing the process currently, you already have a handle on a big piece of the cost, but don’t forget to add in the soft costs of delays in response, management time, contract negotiations, PO processing, travel and other business expenses.
8. Develop an Implementation Project
- One of the first steps should be to define the rules for the selected process. Don’t be surprised if the organization disagrees on the process steps and what the rules are. Be patient and continue working as a team until you have consensus.
- Once the team agrees on the steps and the process rules, work with the users and IT to develop the rules. Make sure you include how to recognize an exception and what the workflow should look like for exceptions.
- Run a batch of transactions or events through the new process and measure the results. Verify that the robot handled each item as expected and that it processed exceptions processed correctly, validate that the right user received the exception notice. Keep running through tests until you are certain that the RPA solution is set up correctly before you cut over.
Using people is too expensive for routine work, and people are happier when they have a variety of engaging tasks to perform. Your business transformation to RPA is the start of the future of work. What are you waiting for?
Most financial institutions realize that the volume of data and analytics required for future success exceeds current processing capabilities. To maximize the potential of machine learning, natural language processing, chatbots, robotic processing automation and intelligent analytics, new technologies will be required.
Subscribe to The Financial Brand via email for FREE!The average bank is drowning in data, from neatly structured numbers to more abstract and hard-to-capture inputs from voice, social media and mobile platforms.
IDC estimates the global generation of data will grow from 16 zettabytes (essentially, 16 trillion gigabytes) to 160 zettabytes in the next ten years, a 30% annual growth clip. And Deloitte forecasts that unstructured data – that hard-to-capture category of data; you can find a primer here – is set to grow at twice that rate annually, with the average financial institution accumulating nine times more unstructured data than structured data by about 2020.
The Reality of Data Overload
The explosive volume of unstructured data that banks are able to process every minute of every day is quickly approaching the point where it can no longer be managed by humans alone. What many banks are realizing is that technology possessing the power to mimic human action and judgment – especially at high speed, scale, quality and lower costs – is necessary in order to keep pace with the looming unstructured data surge on a number of different fronts.
In other words, all of the different technologies that encompass robotic and cognitive automation is fast becoming indispensable necessities to the industry’s data challenge. You’re going to be hearing a lot about this category in the year to come, which includes machine learning, natural language processing, chatbots, robotic processing automation, and intelligent analytics.
The industry’s growing data challenge raises a very important question: Will 2018 be the year of robotic and cognitive automation technologies’ mass adoption by banks big and small?
More Data Requires Greater Automation
The foundation is there for robotic and cognitive automation technology to grow rapidly in the year ahead. It is also being reflected in the marketplace. According to Deloitte’s 2017 “State of Cognitive” survey, 87% of cognitive-aware financial services professionals say that such technologies are important to their products and services, 88% say these technologies are a strategic priority, and just over 35% have invested more than $5 million thus far in such capabilities.
Admittedly, the large, global players in the banking and capital markets sector are in many ways ahead of the curve when it comes to experimenting with, developing and deploying robotics and cognitive solutions. We expect rapid, more democratic adoption across much larger number of banks driven by three factors:
First, banks will increasingly incorporate more information from unstructured data. Regardless of whether a bank has hundreds of thousands or millions of accounts, the rapidly expanding set of unstructured data linked to today’s customers demands that banks will need to develop new muscles to handle that data differently. On a tactical basis, executives will need to evaluate their current processes to determine how to use cognitive technologies to incorporate and sift through the large amount – and different types – of unstructured data.
For instance, banks have historically relied solely on customer-provided data and external sources like credit bureau reports in their account opening process. Today, however, banks must also have more information about an individual or company to affirm an applicant’s identity, sometimes resorting to scouring the Internet or social media for this. This could easily compute to thousands of data points for a single customer.
Second, there is a rapid increase in the level of automation of every bank process. Robotics and cognitive technologies are driving this adoption. Robotics on its own is already well integrated across many banks to complete simple rules-based tasks such as opening email attachments and completing e-forms.
However, the cognitive, analytical element of such tasks is still experimental and siloed. The coming year may be a key turning point in that we are going to see the combined power of robotic and cognitive capabilities become the de facto solution at banks for addressing business process challenges.
Simplification for Improved ROI and a Better Experience
The combination of robotics and cognitive automation could play out in more complex parts of a bank’s business and yield bigger benefits. One such example would be the repairing of payment transactions that currently require manual fixes to remediate issues ranging from the mundane (like sender/receiver information being incomplete) to the highly complex (the payment being a potential fraud case.) If, by combining robotic and cognitive technologies, an average bank could auto-clear even 50% of the original breaks, that could translate into tens of millions of dollar and significantly shorter processing time.
Finally, we believe that automation as a whole will inevitably become transformative for every business process. This will likely begin to play out in 2018. We already are seeing examples of such transformation in pockets — from claims processing completed in seconds, to retail accounts opened in minutes, to loan processing in minutes and hours. Typically, these activities take days or weeks to complete.
No matter the size of your financial institution, the business case for robotic and cognitive automation is robust. Aside from managing dizzying levels of data, it can provide a host of other benefits, including reducing costs, lowering error rates, improving customer churn by providing a markedly higher level of service, increasing the scalability of operations, and improving compliance.
Exploring and adopting these technologies will be critical in order to maintain an edge over competitors in the marketplace and to stay relevant, both next year and in the years to come.