26 October 2017

Why Some RPA Deployments Fail. And What You Can Do About It.

26 October 2017

Why Some RPA Deployments Fail. And What You Can Do About It.

If at first you don’t succeed, try and try again. Initial failures often lead to later success. These are commonly held phrases when it comes to any new or creative endeavor, and though this may seem like an illogical conclusion, these expressions can be no truer when it comes to the the case for robotic process automation (RPA) implementation.

 

For example, a recent report from professional services firm Ernst & Young suggests 30 to 50 percent of RPA projects initially fail. That’s a significant number, but all is not lost for companies looking to implement an automation solution. In fact, professional services firm Genpact softens Ernst & Young’s claim by arguing that:

 

These temporary failures lead to greater learning for an enterprise’s overall digital transformation journey. In today’s hyper-competitive world, the only choice enterprises do not have is to not make a move at all.


Though RPA is widely praised for its ease of implementation, deployments of automation software do not come without challenges. But addressing these challenges results in unforeseen advantages. In order to get a better idea of how RPA deployments can unfold and what can be learned from this process, we’ll examine what can go wrong operationally and how challenges can be exacerbated by the so-called multiplier effect. We’ll also address how failure can be turned into long-term success and how companies can ensure favorable outcomes for their RPA project — from start to finish.

 

Common pitfalls in deploying a RPA solution

 

Of course, integrating or deploying an automation solution has countless numbers of use cases or positive examples across a wide variety of industries, everything from IT services to the Retail industry. Japan-based Nomura Research Institute, however, suggests that: 

 

RPA’s ease of adoption has a downside. Namely, companies tend to decide to implement RPA without cross-organizationally clarifying what they aim to accomplish by doing so and how RPA will fit within the larger scheme of things. As a result, companies are encountering a number of unanticipated difficulties once they start to implement RPA following a successful PoC.

 

As such, some things can also wrong with the deployment of RPA, and a number of unforeseen obstacles can arise.

 

Forgetting business-IT collaboration 

 

A company’s business operations team takes a macro-perspective on long-term goals to carefully select optimized workflows for automation. They closely monitor analytics to ensure that process accuracy, cost reduction, and customer satisfaction benchmarks are met during implementation. A company’s IT team also brings a lot to the table: advanced technical knowledge of existing IT infrastructure, testing and maintenance capabilities, data security oversight, and software support. RPA implementation success is both IT- and business-driven, with RPA governance sitting directly in the space between business and IT. Forgetting to maintain consistent communication between sides will mean that project governance will be weak and that any obstacles, such as potential integration issues of RPA with existing programs, cannot be dealt efficiently.

 

Relying on weak governance

 

As a result of weak governance over RPA projects, many companies underestimate the work required after initial automation is established and think that RPA’s software robots will continue to run autonomously without support. Though RPA never deviates from its configured algorithms, the processes chosen for automation as well as the associated software interfaces and data formats change on a regular, if not weekly, basis.

 

Without a strong governance framework with a clear strategy for the evolution of RPA within the company, it is difficult to ensure that the automation software can be successfully applied to mission-critical processes with the needed scale, accuracy, and speed. In order to avoid automation failures, changes to address the many moving parts of automation have to be continuously planned, communicated, and tested.

 

Automating in a non-efficient way

 

RPA implementation functions at its best when processes have been thoroughly optimized and carefully selected with the involvement of those concerned with the software and the business analysts. When implementing RPA for the first time, however, it’s easy for this step to become an oversight. Companies may quickly get caught up in automating the wrong processes, non-optimized processes, or even too much of a process. RPA is often wrongly used to automate complex, non-standardized processes that require frequent human intervention to complete (think processes with more than 40% exceptions involved). 

 

Building on bad design

 

With the goal of rapid implementation, companies often rely on unstable, non-scalable designs when they initially dive into setting up automation through their RPA deployment. Without a robust framework, however, growing RPA within the company can be difficult, if not futile. Rather than being a one-time event, adopting RPA is a journey — one that, in addition to the software purchase and implementation alone, includes rethinking how it will unfold within a company and change the company over time. Companies must establish a well-designed roadmap that defines the major milestones to success along this journey, including building the required skills, choosing what processes to automate, establishing an RPA budget, selecting an executive sponsorship and running a pilot trial.

 

Though these are a just a few things that can go wrong with RPA deployment, such challenges are not isolated and often arise in groups. Due to weak governance, for example, RPA teams may be forced to rely on an RPA roadmap with bad design. Bad design may also be associated with automating processes that have not yet been optimized for automation. What’s more is that failure of an RPA project is even more likely when such challenges occur simultaneously:

 

Unfortunately, if more than one of these issues occurs – which is common – there’s a significant multiplier effect that can lead to loss of belief in RPA or cause the project to stop, suggests the aforementioned Ernst & Young report.

 

How to ensure successful RPA deployment

 

The statistics on RPA failure are not a reflection of RPA as a technology. Rather, they often reflect a lack of advanced planning, poor project oversight, and a definition of success driven by ROI alone. As such, companies should be aware of the following to ensure that RPA implementation unfolds seamlessly:

 

Advanced planning

 

By establishing a plan for RPA implementation (and for the future of the software within the company) ahead of time, companies will have a stronger footing when it comes time to, for example, integrate the RPA software, onboard employees, and choose which processes to automate.

 

Project oversight

 

Strong governance over the automation project will ensure that RPA deployment happens as planned, especially within the expected time frame and budget, that obstacles can be effectively overcome, and more importantly, that change can be accomodated over time. When systems and applications change–enterprises have intricate and dynamic IT landscapes–the automation needs to remain grounded and unified, and not turn into a fragmentary, hellish operation.

 

Defining success

 

When first implementing RPA, companies should take a broad view on the definition of RPA success. Instead of seeing ROI as the only positive outcome, overcoming RPA challenges and learning as a company on the journey to digital transformation should also be seen as additional measures of success. 

 

Despite potential difficulties during RPA deployment, a recent report by consulting and advisory services company Deloitte argues that automation success can still be easily achievable — especially when a degree of forethought is involved:

 

Implementation challenges can be serious, but proactive planning up front can reduce or eliminate them. [...] Very solvable implementation challenges should not detract from RPA’s ability to open new avenues for organizations and providers to explore, often resulting in large-scale transformation.

 

By fostering a comprehensive understanding of your company’s automation needs and the value proposition RPA provides, you can ensure a successful RPA implementation scheme that is both cost-effective and timely. Here is the comprehensive guide to your RPA journey, created by UiPath in collaboration with Vargha Moyed, EY advisory partner heading the RPA Center of Excellence for EMEA, which details on each and every step you need to take, from pilot to full scale deployment, to ensure success.

 


by Mina Deckard

TOPICS: RPA, RPA challenges, RPA deployment

Show sidebar