Top 5 RPA Pitfalls and How to Avoid Them

Have you recently begun your Robotic Process Automation (RPA) journey, had some bumps along the way, and are now trying to figure out why they occurred and how to get rid of them?

If yes, then this blog is hand-crafted for you.

(If you have already been doing RPA for a longer time, but are experiencing some setbacks – don’t worry. Reading this blog will also give you some cool tips and tricks. I won’t tell anyone about your endeavors, I promise.)

In this blog, I’m going to describe some of the most common pitfalls regarding RPA and help you to understand what the underlying problems are:

• #1: RPA isn’t aligned with strategic objectives
• #2: RPA plans aren't communicated to stakeholders
• #3: Choosing the wrong process for RPA
• #4: Process exceptions aren't caught early enough
• #5: RPA monitoring focuses on bot statistics instead of holistic business performance

You will learn how to avoid these pitfalls, thus be able to reach better ROI for RPA, and take your business performance to the next level.

Watch our on-demand webinar: The 5 RPA Pitfalls and How to Avoid Them

Pitfall #1: RPA isn’t aligned with strategic objectives

Your company has likely set some strategic goals to grow the company, improve business performance, and serve your customers better. In order to reach these goals, companies need to use various tools, and RPA is one of them. This is why RPA should always be closely tied to the strategic objectives of the company.

However, it’s not enough to have your RPA implementations align with your strategy. To get people's buy-in to RPA, the targets must provide impactful outcomes once they are met. In other words, the automated process must have sufficient business impact to get people excited.

Proper RPA targets help manage the stakeholders’ expectations of RPA. This means that the targets must provide enough business impact and be realistic in terms of RPA capabilities. To help with setting such targets I recommend using Process Mining. With process insights provided by process mining you can see where RPA can help you out the most. This also ensures that you’re focusing on the processes that add the most value to your automation investment.

If the word process mining is new to you, be sure to give our "What is Process Mining?" blog a read!

Pitfall #2: RPA plans aren't communicated to stakeholders

Quite often when RPA is implemented, people tend to feel threatened by their new software co-workers. Frankly, this is understandable. Many believe that since RPA bots are much faster than humans at mundane tasks, bots would replace regular employees. Of course, the reality is vastly different. This misunderstanding, however, may lower employees' willingness to co-operate during RPA implementation, which prolongs the implementation time and heightens the costs.

To prevent this from happening, employees should be trained and educated about RPA and its capabilities before implementation. In addition, employees ought to be informed on plans to implement RPA well in advance. These actions help to reduce change resistance and speed up the implementation process. Understanding RPA capabilities not only helps in managing expectations, but also in identifying new automation opportunities.

With data provided by process mining it’s easier to explain why RPA is implemented in the first place. Employees are given an opportunity to see a visual representation of the process' performance to observe. This insight builds acceptance for the implementation. Additionally, you can use your previous successes with RPA to further sell the idea internally.

After seeing results where lead times have been cut in half and dull work has turned more interesting, why wouldn’t employees want to implement RPA as soon as possible? The ability to show proof of previous RPA success is the key to scaling up RPA in your organization.

Pitfall #3: Choosing the wrong process for RPA

Where to implement RPA is perhaps the most important decision to make. It is absolutely critical to select the correct process for automation in order to improve ROI and business buy-in. If you choose the incorrect process, you may be wasting both your time and money as the added value of the RPA won't be sufficient enough. You may not even be able to effectively implement RPA e.g. due to previously unknown exceptions in the process.

Not all processes are suitable for automation. Here’s a list of key process characteristics for most suitable for RPA:

• Standardized & mature process with enough volume
• Simple rule-based tasks that are prone to human error
• High business value (impact!)
• Low cognitive requirements

Process Mining helps you identify the best automation opportunities that align with the above characteristics. You can benchmark automation rates, process standardization, and task volumes across your organization to see where the need for automation is critical.

Benchmarking combined with the ability to test out the business impact of automation will help in creating a business case for RPA. With such insights you will likely find automation opportunities which you never even thought of before!

Pitfall #4: Process exceptions are not caught early enough

Have you experienced delays in RPA implementation due to unexpected process exceptions despite having mapped out the process? This is very common – PwC found that conducting a RPA Proof-of-Concept often takes 4-6 months instead of the expected 4-6 weeks!

The underlying reason behind the delay is in how people map the processes. Typically, exceptions are mapped by interviewing people who work with the process. This activity is not only resource-heavy, but also provides subjectively inaccurate results. As we tend to oversimplify processes, we thus miss half of the exceptions. These undiscovered exceptions play a huge role in delivering RPA on time, as in order to work, RPA needs to be taught how to handle the exceptions in question. If you’re not aware of your process exceptions, how could you teach RPA bots about them?

Unknown exceptions are often encountered after RPA's initial deployment. Finding exceptions this late into the project forces you to jump back and forth from implementation to development, which will delay the project and increase the costs of RPA. Therefore, a better starting point for RPA development would be to completely understand your processes. Such an understanding is possible with process mining: with a data-based analysis all possible variation paths in the process are visible to you.

As such, you do not only save time by skipping lengthy interviews – the results are also accurate. This enables you to develop effective RPA solutions in one go, on time, and within the budget!

Pitfall #5: RPA monitoring focuses on bot statistics instead of holistic business performance

The work with RPA development doesn't end as at the implementation stage. You will also need to be certain that it works as intended. Thus, you want to monitor the performance.

While its nice to look at numbers growing on the screen as RPA completes tasks, you should always make sure that you keep the initial goal in mind. Why did you implement RPA in the first place? Hopefully, the reason was to fulfill a strategic business goal!

If you look through a narrow scope and view only RPA statistics, you won’t see the influence that RPA has on business performance. This implies that you won’t find out how the bot affected process lead times, nor will you see the change in process variations, among other KPIs. Still, such metrics are often used as target measures for RPA implementation. How then could you get a holistic view of how RPA changed your business performance?

Holistic process understanding is at your reach with process mining. Using process mining, you can monitor the effect of RPA on your processes, not just bot statistics. It doesn’t matter what your target for the RPA was. Here, you are able to see how your lead times have shortened and how the automation rates have gone up. With this ability you can instantly find out if RPA is working as you planned. You’re also able to show proof of success to other teams or to your supervisors. Best of all, you can be sure that you reached the goal!

If you’re curious to learn more about RPA powered by process mining, check out our RPA Guide!

Register to our webinar: The 5 RPA Pitfalls and How to Avoid Them