Making RPA Sustainable

Making RPA Sustainable
Making RPA Sustainable

Adhere to three basic principles to ensure the long-term success of your RPA program

Robotic process automation (RPA) has made great strides in fulfilling its promise to the marketplace. RPA is positioned as a no-code end-user computing tool, owned and operated by business functions, that automates routine work tasks with quantifiable business results. Now, businesses have a digital workforce in the form of powerful new end-user computing tools positioned and priced attractively to operations and finance leaders. It has the power to reduce the dependency on traditional automation delivered by IT, improve productivity and accuracy, reduce operational risk, and enable redeployment of expensive human resources to tasks that add more value.

What could be better?

 

However, in practice, though many firms are engaging with RPA, they often initially obtain good results but then stall out and fail to realize RPA’s full potential. Why? What factors prevent RPA from delivering on its promise? Our experience and expertise in implementing RPA leads us to conclude that adhering to three basic principles will ensure that an enterprise is able to achieve and maintain a sustainable RPA approach. The organization must:

Be strategic about where to apply RPA

It is tempting, considering RPA’s presentation-layer integration capability, its agility and its relative user-friendliness, to see RPA as a tactical solution for an organization — one that allows swivel-chair integration among legacy applications, fills in the automation gaps left by all other enterprise applications and ultimately unburdens people from repetitive tasks.

While the return on investment for RPA is easy to justify with tactical implementations, more benefit accrues if the business is able to be more strategic in employing RPA. The risks identified below will also be easier to manage if the business is able to manage RPA as a strategic program.

RPA does all this. However, while the notion of quick wins is valid, it would be a mistake to choose processes based solely on their individual merit. Taking a strategic approach to RPA means that the business processes will be chosen with these synergies in mind and business cases will be established based on the necessary scale and time for deployment. The strategic support of leadership is crucial to benefit fully from RPA. A strategic approach to RPA is required for at least four reasons:

  1. Build strong business cases: Theoretically, a robot replaces approximately five full-time-equivalent (FTE) employees by virtue of working 24/7, processing tasks on average 25 percent faster than a human and eliminating idle time between process steps, especially when handled by multiple people.

Often, business cases are built around the promise of a ratio of at least one robot to three or four FTEs. However, a tactical approach to RPA deployment will fail to deliver such results because to achieve them, the organization must implement enough automated subprocesses to achieve widespread benefits. For instance, what is the benefit of having a tool which can work 24/7 when none of the processes chosen can be performed at night?

  1. RPA scale from simple to complex processes: Initial RPA implementation favors simple processes: a small number of decisions, for example, and fewer than 500 clicks and a limited number of applications accessed. Combine this with large transaction volumes and the business case is easy to make. To ensure RPA sustainability moving forward, organizations must soon find new areas and opportunities, likely including more complex processes, to automate. Fortunately, RPA is getting smarter. By combining RPA maturity and its integration into an organization’s overall digital strategy, new business cases will come into focus. This approach will usher in incremental capabilities such as artificial intelligence (AI).
     
  2. Focus on process standardization and improvement: In many organizations, more often than not, the processes to be automated are not optimal, not fully documented, not fully standardized and sometimes not followed. It is, of course, possible to automate such processes as is. However, this approach likely misses an ideal opportunity to improve performance. To be able to perform automation and conduct process reengineering simultaneously, the RPA deployment team needs to have the strategic mandate to do so, with the backing of top management who accept that the automation effort may last longer but, ultimately, will be much more powerful.
     
  3. Ensure RPA is embedded in technology strategy: Having a tactical approach to RPA deployment runs the risk of missing the strategic importance of RPA from a technological standpoint. RPA changes the business cases for ERP deployment (potentially extending the period between major upgrades), can act as a new integration or backbone for legacy systems, and has the potential to free IT resources from routine API development and maintenance. Understanding this up front will allow for RPA technology to be integrated into the organization’s overall IT strategic roadmap.

Manage the development and delivery risks

In many ways, there is nothing unique about the risk profile of an RPA project that is not equally true of any technology-led implementation. If anything, the relatively noninvasive nature of RPA lowers the risk. However, placing tools with the power of RPA into the hands of individuals without a software-engineering background does change the profile of the risks and the emphasis of some remediation. Similarly, spreadsheets remain valuable tools for most organizations, but creating overly complex spreadsheets, or becoming overly dependent on them, or employing them without the appropriate change management, invites risk.

  1. Consider RPA as a tool to equip the organization to be more agile: RPA was created in response to the frustration of business people in large organizations about what they perceive (rightly or wrongly) as the inertia of their IT colleagues toward pressing business-driven demands.

Implementing RPA requires understanding a process intimately and at its lowest level (i.e., working instructions), along with all its exceptions. With its relative ease of development and user-friendly interface (compared to traditional IT solutions), RPA tilts the balance of required knowledge toward process understanding more than IT. It actually empowers business people to be able to build a required automation relatively quickly.

There are challenges, however, particularly with the need to collaborate with the IT function. At first, many IT professionals manifest what we might call a “not invented here” reaction to RPA. It appears too easy, hence threatening their hard-earned and cherished skills. Rarely have IT professionals been the trigger to start an RPA pilot in an organization. The agile approach of RPA development can make them uncomfortable, and they remain suspicious, at least initially, that RPA could indeed be a noninvasive technology. Thus, by and large, organizations should not expect their IT divisions to be RPA champions and hope that they will be the driving force in pushing for RPA implementation. Another consideration: Because IT groups have the tools and skills to automate activities without RPA, they may not be motivated to implement RPA. However, because other functions in the organization lack the skills to do so, they have little choice but to rely on IT staff for RPA implementation. In these instances, it is beneficial to partner closely with IT to define the business objectives of RPA and outline its benefits and ROI, noting that RPA is unlikely to result in significant displacement of IT resources.

  1. Build the skills required for a full roll-out of RPA: Given that RPA is a relatively new technology, most organizations have limited access to available RPA talent. Fortunately, the basic technology is simple to learn. Someone with some IT background could learn the basics of RPA development with two to three weeks of training; coupled with a few weeks of hands-on practice, this would allow a simple proof of concept to be built. Running a comprehensive RPA program, on the other hand, qualitatively and quantitatively requires another order of magnitude of skills.

In addition, the organization needs to create a team of adequate size to automate enough processes so that the benefits of automation are seen quickly. One effective option is to bring in external help to overcome the initial challenge of an RPA implementation. At the same time, the organization should make arrangements to transfer skills to in-house resources and establish its own center of excellence for RPA implementation and ongoing operation.

  1. Carefully select the best processes to be automated: Once RPA’s potential is demonstrated during a proof of concept, the danger is that the team responsible for the RPA may be overwhelmed by the demand for automation. Requests may be driven by users who wish to automate tasks they do not enjoy and/or by people who have the most power within the organization. This approach might be good from a change-management perspective, but it runs the risk of not delivering the value senior management expects from an RPA program.

Indeed, failing to select processes to be automated carefully and to prioritize them methodically runs the risk of automating those with a low potential for productivity improvement or, worse, choosing processes that use underlying applications scheduled for decommissioning.

To avoid this situation, once the RPA team moves from tactical to strategic approaches, it should confirm it has the right tools to ensure the right decisions are made about priorities, and it should employ a rigorous methodology that will examine processes according to their potential for productivity gain and the complexity of development.

  1. Ensure that you are realistic about how much of the process can be automated: One of the common pitfalls in approaching the use of RPA is wanting to systematically automate 100 percent of a process. This is possible, but generally, 70 to 80 percent is the optimum goal. Trying too hard to cover all the potential exceptions of processes may lead to complex RPA workflows which are time-consuming to develop and hard to maintain or change.

As RPA tools evolve, adding cognitive capabilities based on machine learning, reaching close to 100 percent automation may become achievable. For now, some steps of a process are best left to human intervention. Doing so may require a process redesign, but it ultimately remains a more effective approach.

  1. Be realistic about the return on investment and do not base business cases on the basis of FTE reduction alone: There is a natural tendency to overstate the short-term return on investment (ROI) of an RPA deployment by extrapolating the encouraging results of a pilot, and have it rely solely on time saved (e.g., FTE reduction). Our experience is that the short-term ROI will be less than expected, while the long-term benefits will be more significant.

Furthermore, much more difficult to quantify but nonetheless real is the ability of RPA-enabled resources to be deployed on more value-adding activities and to eliminate human error. Such benefits are potentially more significant than simple FTE reduction, which generally happens only when the enterprise has automated enough processes to reassign human resources to more value-added tasks. Also of note, in shared services centers and/or in situations where employees within a team accept tasks from a queue, RPA’s benefits may be much easier to realize than in business units where automation simply makes an employee’s job easier but does not replace the person.

  1. Include the IT function in the RPA implementation: From both a practical and a strategic perspective, close collaboration with the IT function is crucial when deploying RPA, from the more mundane but crucial factors of infrastructure setup and access rights for robots to more important issues such as application rollouts, changes and decommissioning, all of which can affect the performance of robot.

From a strategic perspective, it is also essential, as argued earlier, for an organization’s IT leaders to fully understand the current and future capabilities of RPA solutions in order to integrate them in the overall IT roadmap. However, while RPA is typically business-driven, it most certainly needs to be governed by the IT function. In order to avoid creating an unsustainable, unmaintainable RPA environment, it is essential to adopt and adhere to testing and development standards. In many cases, IT functions are best placed to provide the necessary oversight. Hence, taking the time to onboard IT professionals is a key success factor in building a sustainable RPA program.

  1. Ensure the effort required to manage stake-holders is fully considered: Rolling out a full RPA program requires the backing of senior management to be able to fund the effort properly and understand its strategic value. Once that is done, promoters also need to have the full collaboration of the IT function (and the cybersecurity functions), as well as the backing of business-line managers who are going to be the beneficiaries of RPA and who need to approve some of the process reengineering and alleviate the concerns of internal audit staff.

“Successfully implementing RPA requires a deep understanding of existing business processes and their suitability for automation. While building a sustainable foundation for RPA is critical when planning for long-term success, understanding an organization’s long-term goals should dictate which RPA solution or platform is best suited.”

-Jonathan Wyatt, Managing Director, Global Head of Protiviti Digital

In addition, to address employee anxiety that deploying RPA will result in layoffs requires a broader change-management and communication program. Furthermore, underestimating the time investment required to manage stakeholders could be fatal to the success of an RPA roll-out, as IT can sabotage the deployment, business-line managers and process owners can refuse to agree with required process changes, and auditors can repeatedly request more information, thus slowing down RPA deployment.

  1. Ensure the delivery method is fit for purpose: Our experience is that organizations tend to apply a traditional software delivery methodology to RPA, requiring low-value documentation tasks and gates, and expecting detailed business requirements, all of which contribute to unnecessary delays in implementation. However, traditional software governance and delivery methods are overengineered for the purposes of RPA. Clearly, organizations should not dismiss all project documentation; they still should record details which enhance quality assurance and those which are crucial for support and maintenance mechanisms. In addition, the enterprise should adopt an agile delivery methodology that allows for increasing levels of automation — for example, the initial implementation cycle may achieve only 50 percent automation, but the goal of a second sprint would be to increase implementation to 70 percent.
     
  2. Ensure a plan is in place to roll out and sustain automation: Organizations should not undertake pilots without developing a roadmap to roll out and sustain RPA, including but not limited to ongoing bot maintenance. Without this, it is easy for RPA initiatives to fail to meet specific requirements and objectives.

Manage the operational risks of achieving the values of RPA

Organizations frequently overlook a number of key operational effects of implementing RPA. These include:

  • Need for supervision;
  • Loss of institutional knowledge;
  • Loss of training platforms; and
  • Costs for maintaining RPA technology.

Following is a brief discussion on each of these effects.

  1. The ongoing need for supervision: Digital operations, like their human counterparts, require supervision: Once the first robot is operational, the need for supervision is ongoing. Supervision involves managing work queues, schedules, execution rules and support, as well as end-to-end visibility for robots, real-time rolling views of robot activity, and clean separation of automation from control and scheduling. Any RPA implementation must take account of the requirement for this new capability and its integration into the existing operating model.
     
  2. Loss of institutional knowledge: Human process knowledge is an undervalued commodity, and once it is programmed into the robot, much of it disappears. With people now gone, the organization is at risk should the need for this knowledge ever reemerge. Governance which includes process documentation and audit trails for decisions made must fill that gap.
     
  3. Loss of a training platform: In many organizations, entry-level employees often, rightly or wrongly, begin their employment tenure engaging in routine tasks and simple processes. When these activities are automated, organizations must rethink their approach to onboarding and training programs.

“The tangible benefits of a properly implemented and executed RPA program are undeniable. However, the RPA ‘buzz’ has led some organizations to take a ‘ready, fire, aim’ approach.”

-Tony Abel, Managing Director, Protiviti

  1. Maintenance costs: Robots must be maintained. Although, by design, robots are resilient (at least to some extent), enterprises should not underestimate the need for adaptive maintenance and its associated cost, which can dilute anticipated ROI. RPA vendors are investing in process analytics and are better recognizing problems and adjusting automatically. However, until this process is fully developed and deployed, maintenance costs will remain a challenge.

In Closing

When performed well, the power of RPA is impressive, as are the results it generates. In order to fully reap the benefits of RPA, however, enterprises must implement it strategically and carefully manage the investment of resources in both RPA development and ongoing operation. Strategic application requires developing strong business cases, progressing from simple to complex processes, and implementing RPA as part of a general process improvement and integrating it into the organization’s overall technology approach.

Furthermore, firms should adopt RPA with the goal of becoming more agile and, in implementing RPA incrementally, should cultivate in-house talent to maintain and expand it, as well as carefully plan which processes should be part of the initial implementation and identify other processes that are candidates for subsequent adoption of RPA. It also is critical to approach RPA implementation with an agile mindset and not treat it akin to a documentation-heavy software-development process. In addition, the organization should secure buy-in not only from the IT and cybersecurity functions, but also from business-line managers, who may resist RPA because they view it as a hindrance to their work rather than a benefit, and the entire workforce, which may perceive RPA as a threat to job security.

Finally, organizations must provide resources for ongoing supervision and maintenance of RPA and consider the implications of the loss of institutional knowledge and elimination of a source of entry-level training when automation supersedes manual processes.

When organizations anticipate and strategically address these factors, they will overcome the challenges of RPA implementation, resulting in the streamlining and simplifying of processes and strategic reallocation of human resources to take on more qualitative, sophisticated responsibilities.

Ultimately, we believe organizations can gain substantial benefits from the use of RPA, even if they do not address every one of the issues we have identified in this paper. However, we believe their results will be more impressive if they take these additional steps.

Contacts

Tony Abel
Managing Director
+1.952.229.2273
[email protected]
Jonathan Wyatt
Managing Director - Global Head of Protiviti Digital
+44.207.024.7522
[email protected]
John Harvie
Director
+44.207.389.0463
[email protected]